diff --git a/poetry.lock b/poetry.lock
index cd87a2d..a5f46f5 100644
--- a/poetry.lock
+++ b/poetry.lock
@@ -29,18 +29,19 @@ dev = ["backports.zoneinfo", "freezegun (>=1.0,<2.0)", "jinja2 (>=3.0)", "pytest
 
 [[package]]
 name = "backrefs"
-version = "5.8"
+version = "5.9"
 description = "A wrapper around re and regex that adds additional back references."
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "backrefs-5.8-py310-none-any.whl", hash = "sha256:c67f6638a34a5b8730812f5101376f9d41dc38c43f1fdc35cb54700f6ed4465d"},
-    {file = "backrefs-5.8-py311-none-any.whl", hash = "sha256:2e1c15e4af0e12e45c8701bd5da0902d326b2e200cafcd25e49d9f06d44bb61b"},
-    {file = "backrefs-5.8-py312-none-any.whl", hash = "sha256:bbef7169a33811080d67cdf1538c8289f76f0942ff971222a16034da88a73486"},
-    {file = "backrefs-5.8-py313-none-any.whl", hash = "sha256:e3a63b073867dbefd0536425f43db618578528e3896fb77be7141328642a1585"},
-    {file = "backrefs-5.8-py39-none-any.whl", hash = "sha256:a66851e4533fb5b371aa0628e1fee1af05135616b86140c9d787a2ffdf4b8fdc"},
-    {file = "backrefs-5.8.tar.gz", hash = "sha256:2cab642a205ce966af3dd4b38ee36009b31fa9502a35fd61d59ccc116e40a6bd"},
+    {file = "backrefs-5.9-py310-none-any.whl", hash = "sha256:db8e8ba0e9de81fcd635f440deab5ae5f2591b54ac1ebe0550a2ca063488cd9f"},
+    {file = "backrefs-5.9-py311-none-any.whl", hash = "sha256:6907635edebbe9b2dc3de3a2befff44d74f30a4562adbb8b36f21252ea19c5cf"},
+    {file = "backrefs-5.9-py312-none-any.whl", hash = "sha256:7fdf9771f63e6028d7fee7e0c497c81abda597ea45d6b8f89e8ad76994f5befa"},
+    {file = "backrefs-5.9-py313-none-any.whl", hash = "sha256:cc37b19fa219e93ff825ed1fed8879e47b4d89aa7a1884860e2db64ccd7c676b"},
+    {file = "backrefs-5.9-py314-none-any.whl", hash = "sha256:df5e169836cc8acb5e440ebae9aad4bf9d15e226d3bad049cf3f6a5c20cc8dc9"},
+    {file = "backrefs-5.9-py39-none-any.whl", hash = "sha256:f48ee18f6252b8f5777a22a00a09a85de0ca931658f1dd96d4406a34f3748c60"},
+    {file = "backrefs-5.9.tar.gz", hash = "sha256:808548cb708d66b82ee231f962cb36faaf4f2baab032f2fbb783e9c2fdddaa59"},
 ]
 
 [package.extras]
@@ -552,67 +553,67 @@ colorama = ">=0.4"
 
 [[package]]
 name = "grpcio"
-version = "1.73.0"
+version = "1.73.1"
 description = "HTTP/2-based RPC framework"
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "grpcio-1.73.0-cp310-cp310-linux_armv7l.whl", hash = "sha256:d050197eeed50f858ef6c51ab09514856f957dba7b1f7812698260fc9cc417f6"},
-    {file = "grpcio-1.73.0-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:ebb8d5f4b0200916fb292a964a4d41210de92aba9007e33d8551d85800ea16cb"},
-    {file = "grpcio-1.73.0-cp310-cp310-manylinux_2_17_aarch64.whl", hash = "sha256:c0811331b469e3f15dda5f90ab71bcd9681189a83944fd6dc908e2c9249041ef"},
-    {file = "grpcio-1.73.0-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:12787c791c3993d0ea1cc8bf90393647e9a586066b3b322949365d2772ba965b"},
-    {file = "grpcio-1.73.0-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2c17771e884fddf152f2a0df12478e8d02853e5b602a10a9a9f1f52fa02b1d32"},
-    {file = "grpcio-1.73.0-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:275e23d4c428c26b51857bbd95fcb8e528783597207ec592571e4372b300a29f"},
-    {file = "grpcio-1.73.0-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:9ffc972b530bf73ef0f948f799482a1bf12d9b6f33406a8e6387c0ca2098a833"},
-    {file = "grpcio-1.73.0-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:ebd8d269df64aff092b2cec5e015d8ae09c7e90888b5c35c24fdca719a2c9f35"},
-    {file = "grpcio-1.73.0-cp310-cp310-win32.whl", hash = "sha256:072d8154b8f74300ed362c01d54af8b93200c1a9077aeaea79828d48598514f1"},
-    {file = "grpcio-1.73.0-cp310-cp310-win_amd64.whl", hash = "sha256:ce953d9d2100e1078a76a9dc2b7338d5415924dc59c69a15bf6e734db8a0f1ca"},
-    {file = "grpcio-1.73.0-cp311-cp311-linux_armv7l.whl", hash = "sha256:51036f641f171eebe5fa7aaca5abbd6150f0c338dab3a58f9111354240fe36ec"},
-    {file = "grpcio-1.73.0-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:d12bbb88381ea00bdd92c55aff3da3391fd85bc902c41275c8447b86f036ce0f"},
-    {file = "grpcio-1.73.0-cp311-cp311-manylinux_2_17_aarch64.whl", hash = "sha256:483c507c2328ed0e01bc1adb13d1eada05cc737ec301d8e5a8f4a90f387f1790"},
-    {file = "grpcio-1.73.0-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c201a34aa960c962d0ce23fe5f423f97e9d4b518ad605eae6d0a82171809caaa"},
-    {file = "grpcio-1.73.0-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:859f70c8e435e8e1fa060e04297c6818ffc81ca9ebd4940e180490958229a45a"},
-    {file = "grpcio-1.73.0-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:e2459a27c6886e7e687e4e407778425f3c6a971fa17a16420227bda39574d64b"},
-    {file = "grpcio-1.73.0-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:e0084d4559ee3dbdcce9395e1bc90fdd0262529b32c417a39ecbc18da8074ac7"},
-    {file = "grpcio-1.73.0-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:ef5fff73d5f724755693a464d444ee0a448c6cdfd3c1616a9223f736c622617d"},
-    {file = "grpcio-1.73.0-cp311-cp311-win32.whl", hash = "sha256:965a16b71a8eeef91fc4df1dc40dc39c344887249174053814f8a8e18449c4c3"},
-    {file = "grpcio-1.73.0-cp311-cp311-win_amd64.whl", hash = "sha256:b71a7b4483d1f753bbc11089ff0f6fa63b49c97a9cc20552cded3fcad466d23b"},
-    {file = "grpcio-1.73.0-cp312-cp312-linux_armv7l.whl", hash = "sha256:fb9d7c27089d9ba3746f18d2109eb530ef2a37452d2ff50f5a6696cd39167d3b"},
-    {file = "grpcio-1.73.0-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:128ba2ebdac41e41554d492b82c34586a90ebd0766f8ebd72160c0e3a57b9155"},
-    {file = "grpcio-1.73.0-cp312-cp312-manylinux_2_17_aarch64.whl", hash = "sha256:068ecc415f79408d57a7f146f54cdf9f0acb4b301a52a9e563973dc981e82f3d"},
-    {file = "grpcio-1.73.0-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:6ddc1cfb2240f84d35d559ade18f69dcd4257dbaa5ba0de1a565d903aaab2968"},
-    {file = "grpcio-1.73.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:e53007f70d9783f53b41b4cf38ed39a8e348011437e4c287eee7dd1d39d54b2f"},
-    {file = "grpcio-1.73.0-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:4dd8d8d092efede7d6f48d695ba2592046acd04ccf421436dd7ed52677a9ad29"},
-    {file = "grpcio-1.73.0-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:70176093d0a95b44d24baa9c034bb67bfe2b6b5f7ebc2836f4093c97010e17fd"},
-    {file = "grpcio-1.73.0-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:085ebe876373ca095e24ced95c8f440495ed0b574c491f7f4f714ff794bbcd10"},
-    {file = "grpcio-1.73.0-cp312-cp312-win32.whl", hash = "sha256:cfc556c1d6aef02c727ec7d0016827a73bfe67193e47c546f7cadd3ee6bf1a60"},
-    {file = "grpcio-1.73.0-cp312-cp312-win_amd64.whl", hash = "sha256:bbf45d59d090bf69f1e4e1594832aaf40aa84b31659af3c5e2c3f6a35202791a"},
-    {file = "grpcio-1.73.0-cp313-cp313-linux_armv7l.whl", hash = "sha256:da1d677018ef423202aca6d73a8d3b2cb245699eb7f50eb5f74cae15a8e1f724"},
-    {file = "grpcio-1.73.0-cp313-cp313-macosx_11_0_universal2.whl", hash = "sha256:36bf93f6a657f37c131d9dd2c391b867abf1426a86727c3575393e9e11dadb0d"},
-    {file = "grpcio-1.73.0-cp313-cp313-manylinux_2_17_aarch64.whl", hash = "sha256:d84000367508ade791d90c2bafbd905574b5ced8056397027a77a215d601ba15"},
-    {file = "grpcio-1.73.0-cp313-cp313-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:c98ba1d928a178ce33f3425ff823318040a2b7ef875d30a0073565e5ceb058d9"},
-    {file = "grpcio-1.73.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:a73c72922dfd30b396a5f25bb3a4590195ee45ecde7ee068acb0892d2900cf07"},
-    {file = "grpcio-1.73.0-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:10e8edc035724aba0346a432060fd192b42bd03675d083c01553cab071a28da5"},
-    {file = "grpcio-1.73.0-cp313-cp313-musllinux_1_1_i686.whl", hash = "sha256:f5cdc332b503c33b1643b12ea933582c7b081957c8bc2ea4cc4bc58054a09288"},
-    {file = "grpcio-1.73.0-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:07ad7c57233c2109e4ac999cb9c2710c3b8e3f491a73b058b0ce431f31ed8145"},
-    {file = "grpcio-1.73.0-cp313-cp313-win32.whl", hash = "sha256:0eb5df4f41ea10bda99a802b2a292d85be28958ede2a50f2beb8c7fc9a738419"},
-    {file = "grpcio-1.73.0-cp313-cp313-win_amd64.whl", hash = "sha256:38cf518cc54cd0c47c9539cefa8888549fcc067db0b0c66a46535ca8032020c4"},
-    {file = "grpcio-1.73.0-cp39-cp39-linux_armv7l.whl", hash = "sha256:1284850607901cfe1475852d808e5a102133461ec9380bc3fc9ebc0686ee8e32"},
-    {file = "grpcio-1.73.0-cp39-cp39-macosx_11_0_universal2.whl", hash = "sha256:0e092a4b28eefb63eec00d09ef33291cd4c3a0875cde29aec4d11d74434d222c"},
-    {file = "grpcio-1.73.0-cp39-cp39-manylinux_2_17_aarch64.whl", hash = "sha256:33577fe7febffe8ebad458744cfee8914e0c10b09f0ff073a6b149a84df8ab8f"},
-    {file = "grpcio-1.73.0-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:60813d8a16420d01fa0da1fc7ebfaaa49a7e5051b0337cd48f4f950eb249a08e"},
-    {file = "grpcio-1.73.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:2a9c957dc65e5d474378d7bcc557e9184576605d4b4539e8ead6e351d7ccce20"},
-    {file = "grpcio-1.73.0-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:3902b71407d021163ea93c70c8531551f71ae742db15b66826cf8825707d2908"},
-    {file = "grpcio-1.73.0-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:1dd7fa7276dcf061e2d5f9316604499eea06b1b23e34a9380572d74fe59915a8"},
-    {file = "grpcio-1.73.0-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:2d1510c4ea473110cb46a010555f2c1a279d1c256edb276e17fa571ba1e8927c"},
-    {file = "grpcio-1.73.0-cp39-cp39-win32.whl", hash = "sha256:d0a1517b2005ba1235a1190b98509264bf72e231215dfeef8db9a5a92868789e"},
-    {file = "grpcio-1.73.0-cp39-cp39-win_amd64.whl", hash = "sha256:6228f7eb6d9f785f38b589d49957fca5df3d5b5349e77d2d89b14e390165344c"},
-    {file = "grpcio-1.73.0.tar.gz", hash = "sha256:3af4c30918a7f0d39de500d11255f8d9da4f30e94a2033e70fe2a720e184bd8e"},
+    {file = "grpcio-1.73.1-cp310-cp310-linux_armv7l.whl", hash = "sha256:2d70f4ddd0a823436c2624640570ed6097e40935c9194482475fe8e3d9754d55"},
+    {file = "grpcio-1.73.1-cp310-cp310-macosx_11_0_universal2.whl", hash = "sha256:3841a8a5a66830261ab6a3c2a3dc539ed84e4ab019165f77b3eeb9f0ba621f26"},
+    {file = "grpcio-1.73.1-cp310-cp310-manylinux_2_17_aarch64.whl", hash = "sha256:628c30f8e77e0258ab788750ec92059fc3d6628590fb4b7cea8c102503623ed7"},
+    {file = "grpcio-1.73.1-cp310-cp310-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:67a0468256c9db6d5ecb1fde4bf409d016f42cef649323f0a08a72f352d1358b"},
+    {file = "grpcio-1.73.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:68b84d65bbdebd5926eb5c53b0b9ec3b3f83408a30e4c20c373c5337b4219ec5"},
+    {file = "grpcio-1.73.1-cp310-cp310-musllinux_1_1_aarch64.whl", hash = "sha256:c54796ca22b8349cc594d18b01099e39f2b7ffb586ad83217655781a350ce4da"},
+    {file = "grpcio-1.73.1-cp310-cp310-musllinux_1_1_i686.whl", hash = "sha256:75fc8e543962ece2f7ecd32ada2d44c0c8570ae73ec92869f9af8b944863116d"},
+    {file = "grpcio-1.73.1-cp310-cp310-musllinux_1_1_x86_64.whl", hash = "sha256:6a6037891cd2b1dd1406b388660522e1565ed340b1fea2955b0234bdd941a862"},
+    {file = "grpcio-1.73.1-cp310-cp310-win32.whl", hash = "sha256:cce7265b9617168c2d08ae570fcc2af4eaf72e84f8c710ca657cc546115263af"},
+    {file = "grpcio-1.73.1-cp310-cp310-win_amd64.whl", hash = "sha256:6a2b372e65fad38842050943f42ce8fee00c6f2e8ea4f7754ba7478d26a356ee"},
+    {file = "grpcio-1.73.1-cp311-cp311-linux_armv7l.whl", hash = "sha256:ba2cea9f7ae4bc21f42015f0ec98f69ae4179848ad744b210e7685112fa507a1"},
+    {file = "grpcio-1.73.1-cp311-cp311-macosx_11_0_universal2.whl", hash = "sha256:d74c3f4f37b79e746271aa6cdb3a1d7e4432aea38735542b23adcabaaee0c097"},
+    {file = "grpcio-1.73.1-cp311-cp311-manylinux_2_17_aarch64.whl", hash = "sha256:5b9b1805a7d61c9e90541cbe8dfe0a593dfc8c5c3a43fe623701b6a01b01d710"},
+    {file = "grpcio-1.73.1-cp311-cp311-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:b3215f69a0670a8cfa2ab53236d9e8026bfb7ead5d4baabe7d7dc11d30fda967"},
+    {file = "grpcio-1.73.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:bc5eccfd9577a5dc7d5612b2ba90cca4ad14c6d949216c68585fdec9848befb1"},
+    {file = "grpcio-1.73.1-cp311-cp311-musllinux_1_1_aarch64.whl", hash = "sha256:dc7d7fd520614fce2e6455ba89791458020a39716951c7c07694f9dbae28e9c0"},
+    {file = "grpcio-1.73.1-cp311-cp311-musllinux_1_1_i686.whl", hash = "sha256:105492124828911f85127e4825d1c1234b032cb9d238567876b5515d01151379"},
+    {file = "grpcio-1.73.1-cp311-cp311-musllinux_1_1_x86_64.whl", hash = "sha256:610e19b04f452ba6f402ac9aa94eb3d21fbc94553368008af634812c4a85a99e"},
+    {file = "grpcio-1.73.1-cp311-cp311-win32.whl", hash = "sha256:d60588ab6ba0ac753761ee0e5b30a29398306401bfbceffe7d68ebb21193f9d4"},
+    {file = "grpcio-1.73.1-cp311-cp311-win_amd64.whl", hash = "sha256:6957025a4608bb0a5ff42abd75bfbb2ed99eda29d5992ef31d691ab54b753643"},
+    {file = "grpcio-1.73.1-cp312-cp312-linux_armv7l.whl", hash = "sha256:921b25618b084e75d424a9f8e6403bfeb7abef074bb6c3174701e0f2542debcf"},
+    {file = "grpcio-1.73.1-cp312-cp312-macosx_11_0_universal2.whl", hash = "sha256:277b426a0ed341e8447fbf6c1d6b68c952adddf585ea4685aa563de0f03df887"},
+    {file = "grpcio-1.73.1-cp312-cp312-manylinux_2_17_aarch64.whl", hash = "sha256:96c112333309493c10e118d92f04594f9055774757f5d101b39f8150f8c25582"},
+    {file = "grpcio-1.73.1-cp312-cp312-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:f48e862aed925ae987eb7084409a80985de75243389dc9d9c271dd711e589918"},
+    {file = "grpcio-1.73.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:83a6c2cce218e28f5040429835fa34a29319071079e3169f9543c3fbeff166d2"},
+    {file = "grpcio-1.73.1-cp312-cp312-musllinux_1_1_aarch64.whl", hash = "sha256:65b0458a10b100d815a8426b1442bd17001fdb77ea13665b2f7dc9e8587fdc6b"},
+    {file = "grpcio-1.73.1-cp312-cp312-musllinux_1_1_i686.whl", hash = "sha256:0a9f3ea8dce9eae9d7cb36827200133a72b37a63896e0e61a9d5ec7d61a59ab1"},
+    {file = "grpcio-1.73.1-cp312-cp312-musllinux_1_1_x86_64.whl", hash = "sha256:de18769aea47f18e782bf6819a37c1c528914bfd5683b8782b9da356506190c8"},
+    {file = "grpcio-1.73.1-cp312-cp312-win32.whl", hash = "sha256:24e06a5319e33041e322d32c62b1e728f18ab8c9dbc91729a3d9f9e3ed336642"},
+    {file = "grpcio-1.73.1-cp312-cp312-win_amd64.whl", hash = "sha256:303c8135d8ab176f8038c14cc10d698ae1db9c480f2b2823f7a987aa2a4c5646"},
+    {file = "grpcio-1.73.1-cp313-cp313-linux_armv7l.whl", hash = "sha256:b310824ab5092cf74750ebd8a8a8981c1810cb2b363210e70d06ef37ad80d4f9"},
+    {file = "grpcio-1.73.1-cp313-cp313-macosx_11_0_universal2.whl", hash = "sha256:8f5a6df3fba31a3485096ac85b2e34b9666ffb0590df0cd044f58694e6a1f6b5"},
+    {file = "grpcio-1.73.1-cp313-cp313-manylinux_2_17_aarch64.whl", hash = "sha256:052e28fe9c41357da42250a91926a3e2f74c046575c070b69659467ca5aa976b"},
+    {file = "grpcio-1.73.1-cp313-cp313-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:1c0bf15f629b1497436596b1cbddddfa3234273490229ca29561209778ebe182"},
+    {file = "grpcio-1.73.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0ab860d5bfa788c5a021fba264802e2593688cd965d1374d31d2b1a34cacd854"},
+    {file = "grpcio-1.73.1-cp313-cp313-musllinux_1_1_aarch64.whl", hash = "sha256:ad1d958c31cc91ab050bd8a91355480b8e0683e21176522bacea225ce51163f2"},
+    {file = "grpcio-1.73.1-cp313-cp313-musllinux_1_1_i686.whl", hash = "sha256:f43ffb3bd415c57224c7427bfb9e6c46a0b6e998754bfa0d00f408e1873dcbb5"},
+    {file = "grpcio-1.73.1-cp313-cp313-musllinux_1_1_x86_64.whl", hash = "sha256:686231cdd03a8a8055f798b2b54b19428cdf18fa1549bee92249b43607c42668"},
+    {file = "grpcio-1.73.1-cp313-cp313-win32.whl", hash = "sha256:89018866a096e2ce21e05eabed1567479713ebe57b1db7cbb0f1e3b896793ba4"},
+    {file = "grpcio-1.73.1-cp313-cp313-win_amd64.whl", hash = "sha256:4a68f8c9966b94dff693670a5cf2b54888a48a5011c5d9ce2295a1a1465ee84f"},
+    {file = "grpcio-1.73.1-cp39-cp39-linux_armv7l.whl", hash = "sha256:b4adc97d2d7f5c660a5498bda978ebb866066ad10097265a5da0511323ae9f50"},
+    {file = "grpcio-1.73.1-cp39-cp39-macosx_11_0_universal2.whl", hash = "sha256:c45a28a0cfb6ddcc7dc50a29de44ecac53d115c3388b2782404218db51cb2df3"},
+    {file = "grpcio-1.73.1-cp39-cp39-manylinux_2_17_aarch64.whl", hash = "sha256:10af9f2ab98a39f5b6c1896c6fc2036744b5b41d12739d48bed4c3e15b6cf900"},
+    {file = "grpcio-1.73.1-cp39-cp39-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:45cf17dcce5ebdb7b4fe9e86cb338fa99d7d1bb71defc78228e1ddf8d0de8cbb"},
+    {file = "grpcio-1.73.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:1c502c2e950fc7e8bf05c047e8a14522ef7babac59abbfde6dbf46b7a0d9c71e"},
+    {file = "grpcio-1.73.1-cp39-cp39-musllinux_1_1_aarch64.whl", hash = "sha256:6abfc0f9153dc4924536f40336f88bd4fe7bd7494f028675e2e04291b8c2c62a"},
+    {file = "grpcio-1.73.1-cp39-cp39-musllinux_1_1_i686.whl", hash = "sha256:ed451a0e39c8e51eb1612b78686839efd1a920666d1666c1adfdb4fd51680c0f"},
+    {file = "grpcio-1.73.1-cp39-cp39-musllinux_1_1_x86_64.whl", hash = "sha256:07f08705a5505c9b5b0cbcbabafb96462b5a15b7236bbf6bbcc6b0b91e1cbd7e"},
+    {file = "grpcio-1.73.1-cp39-cp39-win32.whl", hash = "sha256:ad5c958cc3d98bb9d71714dc69f1c13aaf2f4b53e29d4cc3f1501ef2e4d129b2"},
+    {file = "grpcio-1.73.1-cp39-cp39-win_amd64.whl", hash = "sha256:42f0660bce31b745eb9d23f094a332d31f210dcadd0fc8e5be7e4c62a87ce86b"},
+    {file = "grpcio-1.73.1.tar.gz", hash = "sha256:7fce2cd1c0c1116cf3850564ebfc3264fba75d3c74a7414373f1238ea365ef87"},
 ]
 
 [package.extras]
-protobuf = ["grpcio-tools (>=1.73.0)"]
+protobuf = ["grpcio-tools (>=1.73.1)"]
 
 [[package]]
 name = "idna"
@@ -696,14 +697,14 @@ files = [
 
 [[package]]
 name = "ir-sim"
-version = "2.5.4"
+version = "2.5.5"
 description = "IR-SIM is an open-source, lightweight robot simulator based on Python, designed for robotics navigation, control, and learning. This simulator provides a simple and user-friendly framework for simulating robots, sensors, and environments, facilitating the development and testing of robotics algorithms with minimal hardware requirements."
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "ir_sim-2.5.4-py3-none-any.whl", hash = "sha256:a899f6da22ee53fb0ac150c9d2054d3de86e101e029c775a9265d3b49e9c6885"},
-    {file = "ir_sim-2.5.4.tar.gz", hash = "sha256:30c17a2b1708f9096e19d60968cf7eb77f563008f52ed4b406462c501f76683c"},
+    {file = "ir_sim-2.5.5-py3-none-any.whl", hash = "sha256:2c865da7390594ff939caa561e749d13a6ef2bc29190b3cced46be9b7821ed7d"},
+    {file = "ir_sim-2.5.5.tar.gz", hash = "sha256:fe45ac921ba0e939a19d5f72c250d40721e4391f0f5fb51ce9770e74bcdb0e5f"},
 ]
 
 [package.dependencies]
@@ -1078,14 +1079,14 @@ pyyaml = ">=5.1"
 
 [[package]]
 name = "mkdocs-material"
-version = "9.6.14"
+version = "9.6.15"
 description = "Documentation that simply works"
 optional = false
 python-versions = ">=3.8"
 groups = ["main"]
 files = [
-    {file = "mkdocs_material-9.6.14-py3-none-any.whl", hash = "sha256:3b9cee6d3688551bf7a8e8f41afda97a3c39a12f0325436d76c86706114b721b"},
-    {file = "mkdocs_material-9.6.14.tar.gz", hash = "sha256:39d795e90dce6b531387c255bd07e866e027828b7346d3eba5ac3de265053754"},
+    {file = "mkdocs_material-9.6.15-py3-none-any.whl", hash = "sha256:ac969c94d4fe5eb7c924b6d2f43d7db41159ea91553d18a9afc4780c34f2717a"},
+    {file = "mkdocs_material-9.6.15.tar.gz", hash = "sha256:64adf8fa8dba1a17905b6aee1894a5aafd966d4aeb44a11088519b0f5ca4f1b5"},
 ]
 
 [package.dependencies]
@@ -1529,101 +1530,106 @@ files = [
 
 [[package]]
 name = "pillow"
-version = "11.2.1"
+version = "11.3.0"
 description = "Python Imaging Library (Fork)"
 optional = false
 python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "pillow-11.2.1-cp310-cp310-macosx_10_10_x86_64.whl", hash = "sha256:d57a75d53922fc20c165016a20d9c44f73305e67c351bbc60d1adaf662e74047"},
-    {file = "pillow-11.2.1-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:127bf6ac4a5b58b3d32fc8289656f77f80567d65660bc46f72c0d77e6600cc95"},
-    {file = "pillow-11.2.1-cp310-cp310-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:b4ba4be812c7a40280629e55ae0b14a0aafa150dd6451297562e1764808bbe61"},
-    {file = "pillow-11.2.1-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:c8bd62331e5032bc396a93609982a9ab6b411c05078a52f5fe3cc59234a3abd1"},
-    {file = "pillow-11.2.1-cp310-cp310-manylinux_2_28_aarch64.whl", hash = "sha256:562d11134c97a62fe3af29581f083033179f7ff435f78392565a1ad2d1c2c45c"},
-    {file = "pillow-11.2.1-cp310-cp310-manylinux_2_28_x86_64.whl", hash = "sha256:c97209e85b5be259994eb5b69ff50c5d20cca0f458ef9abd835e262d9d88b39d"},
-    {file = "pillow-11.2.1-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:0c3e6d0f59171dfa2e25d7116217543310908dfa2770aa64b8f87605f8cacc97"},
-    {file = "pillow-11.2.1-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:cc1c3bc53befb6096b84165956e886b1729634a799e9d6329a0c512ab651e579"},
-    {file = "pillow-11.2.1-cp310-cp310-win32.whl", hash = "sha256:312c77b7f07ab2139924d2639860e084ec2a13e72af54d4f08ac843a5fc9c79d"},
-    {file = "pillow-11.2.1-cp310-cp310-win_amd64.whl", hash = "sha256:9bc7ae48b8057a611e5fe9f853baa88093b9a76303937449397899385da06fad"},
-    {file = "pillow-11.2.1-cp310-cp310-win_arm64.whl", hash = "sha256:2728567e249cdd939f6cc3d1f049595c66e4187f3c34078cbc0a7d21c47482d2"},
-    {file = "pillow-11.2.1-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:35ca289f712ccfc699508c4658a1d14652e8033e9b69839edf83cbdd0ba39e70"},
-    {file = "pillow-11.2.1-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:e0409af9f829f87a2dfb7e259f78f317a5351f2045158be321fd135973fff7bf"},
-    {file = "pillow-11.2.1-cp311-cp311-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:d4e5c5edee874dce4f653dbe59db7c73a600119fbea8d31f53423586ee2aafd7"},
-    {file = "pillow-11.2.1-cp311-cp311-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b93a07e76d13bff9444f1a029e0af2964e654bfc2e2c2d46bfd080df5ad5f3d8"},
-    {file = "pillow-11.2.1-cp311-cp311-manylinux_2_28_aarch64.whl", hash = "sha256:e6def7eed9e7fa90fde255afaf08060dc4b343bbe524a8f69bdd2a2f0018f600"},
-    {file = "pillow-11.2.1-cp311-cp311-manylinux_2_28_x86_64.whl", hash = "sha256:8f4f3724c068be008c08257207210c138d5f3731af6c155a81c2b09a9eb3a788"},
-    {file = "pillow-11.2.1-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:a0a6709b47019dff32e678bc12c63008311b82b9327613f534e496dacaefb71e"},
-    {file = "pillow-11.2.1-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:f6b0c664ccb879109ee3ca702a9272d877f4fcd21e5eb63c26422fd6e415365e"},
-    {file = "pillow-11.2.1-cp311-cp311-win32.whl", hash = "sha256:cc5d875d56e49f112b6def6813c4e3d3036d269c008bf8aef72cd08d20ca6df6"},
-    {file = "pillow-11.2.1-cp311-cp311-win_amd64.whl", hash = "sha256:0f5c7eda47bf8e3c8a283762cab94e496ba977a420868cb819159980b6709193"},
-    {file = "pillow-11.2.1-cp311-cp311-win_arm64.whl", hash = "sha256:4d375eb838755f2528ac8cbc926c3e31cc49ca4ad0cf79cff48b20e30634a4a7"},
-    {file = "pillow-11.2.1-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:78afba22027b4accef10dbd5eed84425930ba41b3ea0a86fa8d20baaf19d807f"},
-    {file = "pillow-11.2.1-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:78092232a4ab376a35d68c4e6d5e00dfd73454bd12b230420025fbe178ee3b0b"},
-    {file = "pillow-11.2.1-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:25a5f306095c6780c52e6bbb6109624b95c5b18e40aab1c3041da3e9e0cd3e2d"},
-    {file = "pillow-11.2.1-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:0c7b29dbd4281923a2bfe562acb734cee96bbb129e96e6972d315ed9f232bef4"},
-    {file = "pillow-11.2.1-cp312-cp312-manylinux_2_28_aarch64.whl", hash = "sha256:3e645b020f3209a0181a418bffe7b4a93171eef6c4ef6cc20980b30bebf17b7d"},
-    {file = "pillow-11.2.1-cp312-cp312-manylinux_2_28_x86_64.whl", hash = "sha256:b2dbea1012ccb784a65349f57bbc93730b96e85b42e9bf7b01ef40443db720b4"},
-    {file = "pillow-11.2.1-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:da3104c57bbd72948d75f6a9389e6727d2ab6333c3617f0a89d72d4940aa0443"},
-    {file = "pillow-11.2.1-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:598174aef4589af795f66f9caab87ba4ff860ce08cd5bb447c6fc553ffee603c"},
-    {file = "pillow-11.2.1-cp312-cp312-win32.whl", hash = "sha256:1d535df14716e7f8776b9e7fee118576d65572b4aad3ed639be9e4fa88a1cad3"},
-    {file = "pillow-11.2.1-cp312-cp312-win_amd64.whl", hash = "sha256:14e33b28bf17c7a38eede290f77db7c664e4eb01f7869e37fa98a5aa95978941"},
-    {file = "pillow-11.2.1-cp312-cp312-win_arm64.whl", hash = "sha256:21e1470ac9e5739ff880c211fc3af01e3ae505859392bf65458c224d0bf283eb"},
-    {file = "pillow-11.2.1-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:fdec757fea0b793056419bca3e9932eb2b0ceec90ef4813ea4c1e072c389eb28"},
-    {file = "pillow-11.2.1-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:b0e130705d568e2f43a17bcbe74d90958e8a16263868a12c3e0d9c8162690830"},
-    {file = "pillow-11.2.1-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:7bdb5e09068332578214cadd9c05e3d64d99e0e87591be22a324bdbc18925be0"},
-    {file = "pillow-11.2.1-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:d189ba1bebfbc0c0e529159631ec72bb9e9bc041f01ec6d3233d6d82eb823bc1"},
-    {file = "pillow-11.2.1-cp313-cp313-manylinux_2_28_aarch64.whl", hash = "sha256:191955c55d8a712fab8934a42bfefbf99dd0b5875078240943f913bb66d46d9f"},
-    {file = "pillow-11.2.1-cp313-cp313-manylinux_2_28_x86_64.whl", hash = "sha256:ad275964d52e2243430472fc5d2c2334b4fc3ff9c16cb0a19254e25efa03a155"},
-    {file = "pillow-11.2.1-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:750f96efe0597382660d8b53e90dd1dd44568a8edb51cb7f9d5d918b80d4de14"},
-    {file = "pillow-11.2.1-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:fe15238d3798788d00716637b3d4e7bb6bde18b26e5d08335a96e88564a36b6b"},
-    {file = "pillow-11.2.1-cp313-cp313-win32.whl", hash = "sha256:3fe735ced9a607fee4f481423a9c36701a39719252a9bb251679635f99d0f7d2"},
-    {file = "pillow-11.2.1-cp313-cp313-win_amd64.whl", hash = "sha256:74ee3d7ecb3f3c05459ba95eed5efa28d6092d751ce9bf20e3e253a4e497e691"},
-    {file = "pillow-11.2.1-cp313-cp313-win_arm64.whl", hash = "sha256:5119225c622403afb4b44bad4c1ca6c1f98eed79db8d3bc6e4e160fc6339d66c"},
-    {file = "pillow-11.2.1-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:8ce2e8411c7aaef53e6bb29fe98f28cd4fbd9a1d9be2eeea434331aac0536b22"},
-    {file = "pillow-11.2.1-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:9ee66787e095127116d91dea2143db65c7bb1e232f617aa5957c0d9d2a3f23a7"},
-    {file = "pillow-11.2.1-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:9622e3b6c1d8b551b6e6f21873bdcc55762b4b2126633014cea1803368a9aa16"},
-    {file = "pillow-11.2.1-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:63b5dff3a68f371ea06025a1a6966c9a1e1ee452fc8020c2cd0ea41b83e9037b"},
-    {file = "pillow-11.2.1-cp313-cp313t-manylinux_2_28_aarch64.whl", hash = "sha256:31df6e2d3d8fc99f993fd253e97fae451a8db2e7207acf97859732273e108406"},
-    {file = "pillow-11.2.1-cp313-cp313t-manylinux_2_28_x86_64.whl", hash = "sha256:062b7a42d672c45a70fa1f8b43d1d38ff76b63421cbbe7f88146b39e8a558d91"},
-    {file = "pillow-11.2.1-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:4eb92eca2711ef8be42fd3f67533765d9fd043b8c80db204f16c8ea62ee1a751"},
-    {file = "pillow-11.2.1-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:f91ebf30830a48c825590aede79376cb40f110b387c17ee9bd59932c961044f9"},
-    {file = "pillow-11.2.1-cp313-cp313t-win32.whl", hash = "sha256:e0b55f27f584ed623221cfe995c912c61606be8513bfa0e07d2c674b4516d9dd"},
-    {file = "pillow-11.2.1-cp313-cp313t-win_amd64.whl", hash = "sha256:36d6b82164c39ce5482f649b437382c0fb2395eabc1e2b1702a6deb8ad647d6e"},
-    {file = "pillow-11.2.1-cp313-cp313t-win_arm64.whl", hash = "sha256:225c832a13326e34f212d2072982bb1adb210e0cc0b153e688743018c94a2681"},
-    {file = "pillow-11.2.1-cp39-cp39-macosx_10_10_x86_64.whl", hash = "sha256:7491cf8a79b8eb867d419648fff2f83cb0b3891c8b36da92cc7f1931d46108c8"},
-    {file = "pillow-11.2.1-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:8b02d8f9cb83c52578a0b4beadba92e37d83a4ef11570a8688bbf43f4ca50909"},
-    {file = "pillow-11.2.1-cp39-cp39-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:014ca0050c85003620526b0ac1ac53f56fc93af128f7546623cc8e31875ab928"},
-    {file = "pillow-11.2.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3692b68c87096ac6308296d96354eddd25f98740c9d2ab54e1549d6c8aea9d79"},
-    {file = "pillow-11.2.1-cp39-cp39-manylinux_2_28_aarch64.whl", hash = "sha256:f781dcb0bc9929adc77bad571b8621ecb1e4cdef86e940fe2e5b5ee24fd33b35"},
-    {file = "pillow-11.2.1-cp39-cp39-manylinux_2_28_x86_64.whl", hash = "sha256:2b490402c96f907a166615e9a5afacf2519e28295f157ec3a2bb9bd57de638cb"},
-    {file = "pillow-11.2.1-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:dd6b20b93b3ccc9c1b597999209e4bc5cf2853f9ee66e3fc9a400a78733ffc9a"},
-    {file = "pillow-11.2.1-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:4b835d89c08a6c2ee7781b8dd0a30209a8012b5f09c0a665b65b0eb3560b6f36"},
-    {file = "pillow-11.2.1-cp39-cp39-win32.whl", hash = "sha256:b10428b3416d4f9c61f94b494681280be7686bda15898a3a9e08eb66a6d92d67"},
-    {file = "pillow-11.2.1-cp39-cp39-win_amd64.whl", hash = "sha256:6ebce70c3f486acf7591a3d73431fa504a4e18a9b97ff27f5f47b7368e4b9dd1"},
-    {file = "pillow-11.2.1-cp39-cp39-win_arm64.whl", hash = "sha256:c27476257b2fdcd7872d54cfd119b3a9ce4610fb85c8e32b70b42e3680a29a1e"},
-    {file = "pillow-11.2.1-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:9b7b0d4fd2635f54ad82785d56bc0d94f147096493a79985d0ab57aedd563156"},
-    {file = "pillow-11.2.1-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:aa442755e31c64037aa7c1cb186e0b369f8416c567381852c63444dd666fb772"},
-    {file = "pillow-11.2.1-pp310-pypy310_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:f0d3348c95b766f54b76116d53d4cb171b52992a1027e7ca50c81b43b9d9e363"},
-    {file = "pillow-11.2.1-pp310-pypy310_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:85d27ea4c889342f7e35f6d56e7e1cb345632ad592e8c51b693d7b7556043ce0"},
-    {file = "pillow-11.2.1-pp310-pypy310_pp73-manylinux_2_28_aarch64.whl", hash = "sha256:bf2c33d6791c598142f00c9c4c7d47f6476731c31081331664eb26d6ab583e01"},
-    {file = "pillow-11.2.1-pp310-pypy310_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:e616e7154c37669fc1dfc14584f11e284e05d1c650e1c0f972f281c4ccc53193"},
-    {file = "pillow-11.2.1-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:39ad2e0f424394e3aebc40168845fee52df1394a4673a6ee512d840d14ab3013"},
-    {file = "pillow-11.2.1-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:80f1df8dbe9572b4b7abdfa17eb5d78dd620b1d55d9e25f834efdbee872d3aed"},
-    {file = "pillow-11.2.1-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:ea926cfbc3957090becbcbbb65ad177161a2ff2ad578b5a6ec9bb1e1cd78753c"},
-    {file = "pillow-11.2.1-pp311-pypy311_pp73-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:738db0e0941ca0376804d4de6a782c005245264edaa253ffce24e5a15cbdc7bd"},
-    {file = "pillow-11.2.1-pp311-pypy311_pp73-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:9db98ab6565c69082ec9b0d4e40dd9f6181dab0dd236d26f7a50b8b9bfbd5076"},
-    {file = "pillow-11.2.1-pp311-pypy311_pp73-manylinux_2_28_aarch64.whl", hash = "sha256:036e53f4170e270ddb8797d4c590e6dd14d28e15c7da375c18978045f7e6c37b"},
-    {file = "pillow-11.2.1-pp311-pypy311_pp73-manylinux_2_28_x86_64.whl", hash = "sha256:14f73f7c291279bd65fda51ee87affd7c1e097709f7fdd0188957a16c264601f"},
-    {file = "pillow-11.2.1-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:208653868d5c9ecc2b327f9b9ef34e0e42a4cdd172c2988fd81d62d2bc9bc044"},
-    {file = "pillow-11.2.1.tar.gz", hash = "sha256:a64dd61998416367b7ef979b73d3a85853ba9bec4c2925f74e588879a58716b6"},
+    {file = "pillow-11.3.0-cp310-cp310-macosx_10_10_x86_64.whl", hash = "sha256:1b9c17fd4ace828b3003dfd1e30bff24863e0eb59b535e8f80194d9cc7ecf860"},
+    {file = "pillow-11.3.0-cp310-cp310-macosx_11_0_arm64.whl", hash = "sha256:65dc69160114cdd0ca0f35cb434633c75e8e7fad4cf855177a05bf38678f73ad"},
+    {file = "pillow-11.3.0-cp310-cp310-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:f1f182ebd2303acf8c380a54f615ec883322593320a9b00438eb842c1f37ae50"},
+    {file = "pillow-11.3.0-cp310-cp310-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:4445fa62e15936a028672fd48c4c11a66d641d2c05726c7ec1f8ba6a572036ae"},
+    {file = "pillow-11.3.0-cp310-cp310-musllinux_1_2_aarch64.whl", hash = "sha256:71f511f6b3b91dd543282477be45a033e4845a40278fa8dcdbfdb07109bf18f9"},
+    {file = "pillow-11.3.0-cp310-cp310-musllinux_1_2_x86_64.whl", hash = "sha256:040a5b691b0713e1f6cbe222e0f4f74cd233421e105850ae3b3c0ceda520f42e"},
+    {file = "pillow-11.3.0-cp310-cp310-win32.whl", hash = "sha256:89bd777bc6624fe4115e9fac3352c79ed60f3bb18651420635f26e643e3dd1f6"},
+    {file = "pillow-11.3.0-cp310-cp310-win_amd64.whl", hash = "sha256:19d2ff547c75b8e3ff46f4d9ef969a06c30ab2d4263a9e287733aa8b2429ce8f"},
+    {file = "pillow-11.3.0-cp310-cp310-win_arm64.whl", hash = "sha256:819931d25e57b513242859ce1876c58c59dc31587847bf74cfe06b2e0cb22d2f"},
+    {file = "pillow-11.3.0-cp311-cp311-macosx_10_10_x86_64.whl", hash = "sha256:1cd110edf822773368b396281a2293aeb91c90a2db00d78ea43e7e861631b722"},
+    {file = "pillow-11.3.0-cp311-cp311-macosx_11_0_arm64.whl", hash = "sha256:9c412fddd1b77a75aa904615ebaa6001f169b26fd467b4be93aded278266b288"},
+    {file = "pillow-11.3.0-cp311-cp311-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:643f189248837533073c405ec2f0bb250ba54598cf80e8c1e043381a60632f58"},
+    {file = "pillow-11.3.0-cp311-cp311-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:106064daa23a745510dabce1d84f29137a37224831d88eb4ce94bb187b1d7e5f"},
+    {file = "pillow-11.3.0-cp311-cp311-musllinux_1_2_aarch64.whl", hash = "sha256:cd8ff254faf15591e724dc7c4ddb6bf4793efcbe13802a4ae3e863cd300b493e"},
+    {file = "pillow-11.3.0-cp311-cp311-musllinux_1_2_x86_64.whl", hash = "sha256:932c754c2d51ad2b2271fd01c3d121daaa35e27efae2a616f77bf164bc0b3e94"},
+    {file = "pillow-11.3.0-cp311-cp311-win32.whl", hash = "sha256:b4b8f3efc8d530a1544e5962bd6b403d5f7fe8b9e08227c6b255f98ad82b4ba0"},
+    {file = "pillow-11.3.0-cp311-cp311-win_amd64.whl", hash = "sha256:1a992e86b0dd7aeb1f053cd506508c0999d710a8f07b4c791c63843fc6a807ac"},
+    {file = "pillow-11.3.0-cp311-cp311-win_arm64.whl", hash = "sha256:30807c931ff7c095620fe04448e2c2fc673fcbb1ffe2a7da3fb39613489b1ddd"},
+    {file = "pillow-11.3.0-cp312-cp312-macosx_10_13_x86_64.whl", hash = "sha256:fdae223722da47b024b867c1ea0be64e0df702c5e0a60e27daad39bf960dd1e4"},
+    {file = "pillow-11.3.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:921bd305b10e82b4d1f5e802b6850677f965d8394203d182f078873851dada69"},
+    {file = "pillow-11.3.0-cp312-cp312-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:97f07ed9f56a3b9b5f49d3661dc9607484e85c67e27f3e8be2c7d28ca032fec7"},
+    {file = "pillow-11.3.0-cp312-cp312-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:676b2815362456b5b3216b4fd5bd89d362100dc6f4945154ff172e206a22c024"},
+    {file = "pillow-11.3.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:3e184b2f26ff146363dd07bde8b711833d7b0202e27d13540bfe2e35a323a809"},
+    {file = "pillow-11.3.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:6be31e3fc9a621e071bc17bb7de63b85cbe0bfae91bb0363c893cbe67247780d"},
+    {file = "pillow-11.3.0-cp312-cp312-win32.whl", hash = "sha256:7b161756381f0918e05e7cb8a371fff367e807770f8fe92ecb20d905d0e1c149"},
+    {file = "pillow-11.3.0-cp312-cp312-win_amd64.whl", hash = "sha256:a6444696fce635783440b7f7a9fc24b3ad10a9ea3f0ab66c5905be1c19ccf17d"},
+    {file = "pillow-11.3.0-cp312-cp312-win_arm64.whl", hash = "sha256:2aceea54f957dd4448264f9bf40875da0415c83eb85f55069d89c0ed436e3542"},
+    {file = "pillow-11.3.0-cp313-cp313-ios_13_0_arm64_iphoneos.whl", hash = "sha256:1c627742b539bba4309df89171356fcb3cc5a9178355b2727d1b74a6cf155fbd"},
+    {file = "pillow-11.3.0-cp313-cp313-ios_13_0_arm64_iphonesimulator.whl", hash = "sha256:30b7c02f3899d10f13d7a48163c8969e4e653f8b43416d23d13d1bbfdc93b9f8"},
+    {file = "pillow-11.3.0-cp313-cp313-ios_13_0_x86_64_iphonesimulator.whl", hash = "sha256:7859a4cc7c9295f5838015d8cc0a9c215b77e43d07a25e460f35cf516df8626f"},
+    {file = "pillow-11.3.0-cp313-cp313-macosx_10_13_x86_64.whl", hash = "sha256:ec1ee50470b0d050984394423d96325b744d55c701a439d2bd66089bff963d3c"},
+    {file = "pillow-11.3.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:7db51d222548ccfd274e4572fdbf3e810a5e66b00608862f947b163e613b67dd"},
+    {file = "pillow-11.3.0-cp313-cp313-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:c37d8ba9411d6003bba9e518db0db0c58a680ab9fe5179f040b0463644bc9805"},
+    {file = "pillow-11.3.0-cp313-cp313-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:13f87d581e71d9189ab21fe0efb5a23e9f28552d5be6979e84001d3b8505abe8"},
+    {file = "pillow-11.3.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:023f6d2d11784a465f09fd09a34b150ea4672e85fb3d05931d89f373ab14abb2"},
+    {file = "pillow-11.3.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:45dfc51ac5975b938e9809451c51734124e73b04d0f0ac621649821a63852e7b"},
+    {file = "pillow-11.3.0-cp313-cp313-win32.whl", hash = "sha256:a4d336baed65d50d37b88ca5b60c0fa9d81e3a87d4a7930d3880d1624d5b31f3"},
+    {file = "pillow-11.3.0-cp313-cp313-win_amd64.whl", hash = "sha256:0bce5c4fd0921f99d2e858dc4d4d64193407e1b99478bc5cacecba2311abde51"},
+    {file = "pillow-11.3.0-cp313-cp313-win_arm64.whl", hash = "sha256:1904e1264881f682f02b7f8167935cce37bc97db457f8e7849dc3a6a52b99580"},
+    {file = "pillow-11.3.0-cp313-cp313t-macosx_10_13_x86_64.whl", hash = "sha256:4c834a3921375c48ee6b9624061076bc0a32a60b5532b322cc0ea64e639dd50e"},
+    {file = "pillow-11.3.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:5e05688ccef30ea69b9317a9ead994b93975104a677a36a8ed8106be9260aa6d"},
+    {file = "pillow-11.3.0-cp313-cp313t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:1f85acb69adf2aaee8b7da124efebbdb959a104db34d3a2cb0f3793dbae422a8"},
+    {file = "pillow-11.3.0-cp313-cp313t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:05f6ecbeff5005399bb48d198f098a9b4b6bdf27b8487c7f38ca16eeb070cd59"},
+    {file = "pillow-11.3.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:a7bc6e6fd0395bc052f16b1a8670859964dbd7003bd0af2ff08342eb6e442cfe"},
+    {file = "pillow-11.3.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:83e1b0161c9d148125083a35c1c5a89db5b7054834fd4387499e06552035236c"},
+    {file = "pillow-11.3.0-cp313-cp313t-win32.whl", hash = "sha256:2a3117c06b8fb646639dce83694f2f9eac405472713fcb1ae887469c0d4f6788"},
+    {file = "pillow-11.3.0-cp313-cp313t-win_amd64.whl", hash = "sha256:857844335c95bea93fb39e0fa2726b4d9d758850b34075a7e3ff4f4fa3aa3b31"},
+    {file = "pillow-11.3.0-cp313-cp313t-win_arm64.whl", hash = "sha256:8797edc41f3e8536ae4b10897ee2f637235c94f27404cac7297f7b607dd0716e"},
+    {file = "pillow-11.3.0-cp314-cp314-macosx_10_13_x86_64.whl", hash = "sha256:d9da3df5f9ea2a89b81bb6087177fb1f4d1c7146d583a3fe5c672c0d94e55e12"},
+    {file = "pillow-11.3.0-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:0b275ff9b04df7b640c59ec5a3cb113eefd3795a8df80bac69646ef699c6981a"},
+    {file = "pillow-11.3.0-cp314-cp314-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:41742638139424703b4d01665b807c6468e23e699e8e90cffefe291c5832b027"},
+    {file = "pillow-11.3.0-cp314-cp314-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:93efb0b4de7e340d99057415c749175e24c8864302369e05914682ba642e5d77"},
+    {file = "pillow-11.3.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:7966e38dcd0fa11ca390aed7c6f20454443581d758242023cf36fcb319b1a874"},
+    {file = "pillow-11.3.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:98a9afa7b9007c67ed84c57c9e0ad86a6000da96eaa638e4f8abe5b65ff83f0a"},
+    {file = "pillow-11.3.0-cp314-cp314-win32.whl", hash = "sha256:02a723e6bf909e7cea0dac1b0e0310be9d7650cd66222a5f1c571455c0a45214"},
+    {file = "pillow-11.3.0-cp314-cp314-win_amd64.whl", hash = "sha256:a418486160228f64dd9e9efcd132679b7a02a5f22c982c78b6fc7dab3fefb635"},
+    {file = "pillow-11.3.0-cp314-cp314-win_arm64.whl", hash = "sha256:155658efb5e044669c08896c0c44231c5e9abcaadbc5cd3648df2f7c0b96b9a6"},
+    {file = "pillow-11.3.0-cp314-cp314t-macosx_10_13_x86_64.whl", hash = "sha256:59a03cdf019efbfeeed910bf79c7c93255c3d54bc45898ac2a4140071b02b4ae"},
+    {file = "pillow-11.3.0-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:f8a5827f84d973d8636e9dc5764af4f0cf2318d26744b3d902931701b0d46653"},
+    {file = "pillow-11.3.0-cp314-cp314t-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:4c96f993ab8c98460cd0c001447bff6194403e8b1d7e149ade5f00594918128b"},
+    {file = "pillow-11.3.0-cp314-cp314t-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:41342b64afeba938edb034d122b2dda5db2139b9a4af999729ba8818e0056477"},
+    {file = "pillow-11.3.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:068d9c39a2d1b358eb9f245ce7ab1b5c3246c7c8c7d9ba58cfa5b43146c06e50"},
+    {file = "pillow-11.3.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:a1bc6ba083b145187f648b667e05a2534ecc4b9f2784c2cbe3089e44868f2b9b"},
+    {file = "pillow-11.3.0-cp314-cp314t-win32.whl", hash = "sha256:118ca10c0d60b06d006be10a501fd6bbdfef559251ed31b794668ed569c87e12"},
+    {file = "pillow-11.3.0-cp314-cp314t-win_amd64.whl", hash = "sha256:8924748b688aa210d79883357d102cd64690e56b923a186f35a82cbc10f997db"},
+    {file = "pillow-11.3.0-cp314-cp314t-win_arm64.whl", hash = "sha256:79ea0d14d3ebad43ec77ad5272e6ff9bba5b679ef73375ea760261207fa8e0aa"},
+    {file = "pillow-11.3.0-cp39-cp39-macosx_10_10_x86_64.whl", hash = "sha256:48d254f8a4c776de343051023eb61ffe818299eeac478da55227d96e241de53f"},
+    {file = "pillow-11.3.0-cp39-cp39-macosx_11_0_arm64.whl", hash = "sha256:7aee118e30a4cf54fdd873bd3a29de51e29105ab11f9aad8c32123f58c8f8081"},
+    {file = "pillow-11.3.0-cp39-cp39-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:092c80c76635f5ecb10f3f83d76716165c96f5229addbd1ec2bdbbda7d496e06"},
+    {file = "pillow-11.3.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:cadc9e0ea0a2431124cde7e1697106471fc4c1da01530e679b2391c37d3fbb3a"},
+    {file = "pillow-11.3.0-cp39-cp39-musllinux_1_2_aarch64.whl", hash = "sha256:6a418691000f2a418c9135a7cf0d797c1bb7d9a485e61fe8e7722845b95ef978"},
+    {file = "pillow-11.3.0-cp39-cp39-musllinux_1_2_x86_64.whl", hash = "sha256:97afb3a00b65cc0804d1c7abddbf090a81eaac02768af58cbdcaaa0a931e0b6d"},
+    {file = "pillow-11.3.0-cp39-cp39-win32.whl", hash = "sha256:ea944117a7974ae78059fcc1800e5d3295172bb97035c0c1d9345fca1419da71"},
+    {file = "pillow-11.3.0-cp39-cp39-win_amd64.whl", hash = "sha256:e5c5858ad8ec655450a7c7df532e9842cf8df7cc349df7225c60d5d348c8aada"},
+    {file = "pillow-11.3.0-cp39-cp39-win_arm64.whl", hash = "sha256:6abdbfd3aea42be05702a8dd98832329c167ee84400a1d1f61ab11437f1717eb"},
+    {file = "pillow-11.3.0-pp310-pypy310_pp73-macosx_10_15_x86_64.whl", hash = "sha256:3cee80663f29e3843b68199b9d6f4f54bd1d4a6b59bdd91bceefc51238bcb967"},
+    {file = "pillow-11.3.0-pp310-pypy310_pp73-macosx_11_0_arm64.whl", hash = "sha256:b5f56c3f344f2ccaf0dd875d3e180f631dc60a51b314295a3e681fe8cf851fbe"},
+    {file = "pillow-11.3.0-pp310-pypy310_pp73-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:527b37216b6ac3a12d7838dc3bd75208ec57c1c6d11ef01902266a5a0c14fc27"},
+    {file = "pillow-11.3.0-pp310-pypy310_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:be5463ac478b623b9dd3937afd7fb7ab3d79dd290a28e2b6df292dc75063eb8a"},
+    {file = "pillow-11.3.0-pp310-pypy310_pp73-win_amd64.whl", hash = "sha256:8dc70ca24c110503e16918a658b869019126ecfe03109b754c402daff12b3d9f"},
+    {file = "pillow-11.3.0-pp311-pypy311_pp73-macosx_10_15_x86_64.whl", hash = "sha256:7c8ec7a017ad1bd562f93dbd8505763e688d388cde6e4a010ae1486916e713e6"},
+    {file = "pillow-11.3.0-pp311-pypy311_pp73-macosx_11_0_arm64.whl", hash = "sha256:9ab6ae226de48019caa8074894544af5b53a117ccb9d3b3dcb2871464c829438"},
+    {file = "pillow-11.3.0-pp311-pypy311_pp73-manylinux_2_27_aarch64.manylinux_2_28_aarch64.whl", hash = "sha256:5418b53c0d59b3824d05e029669efa023bbef0f3e92e75ec8428f3799487f361"},
+    {file = "pillow-11.3.0-pp311-pypy311_pp73-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl", hash = "sha256:504b6f59505f08ae014f724b6207ff6222662aab5cc9542577fb084ed0676ac7"},
+    {file = "pillow-11.3.0-pp311-pypy311_pp73-win_amd64.whl", hash = "sha256:c84d689db21a1c397d001aa08241044aa2069e7587b398c8cc63020390b1c1b8"},
+    {file = "pillow-11.3.0.tar.gz", hash = "sha256:3828ee7586cd0b2091b6209e5ad53e20d0649bbe87164a459d0676e035e8f523"},
 ]
 
 [package.extras]
-docs = ["furo", "olefile", "sphinx (>=8.2)", "sphinx-copybutton", "sphinx-inline-tabs", "sphinxext-opengraph"]
+docs = ["furo", "olefile", "sphinx (>=8.2)", "sphinx-autobuild", "sphinx-copybutton", "sphinx-inline-tabs", "sphinxext-opengraph"]
 fpx = ["olefile"]
 mic = ["olefile"]
 test-arrow = ["pyarrow"]
-tests = ["check-manifest", "coverage (>=7.4.2)", "defusedxml", "markdown2", "olefile", "packaging", "pyroma", "pytest", "pytest-cov", "pytest-timeout", "trove-classifiers (>=2024.10.12)"]
+tests = ["check-manifest", "coverage (>=7.4.2)", "defusedxml", "markdown2", "olefile", "packaging", "pyroma", "pytest", "pytest-cov", "pytest-timeout", "pytest-xdist", "trove-classifiers (>=2024.10.12)"]
 typing = ["typing-extensions"]
 xmp = ["defusedxml"]
 
@@ -1705,14 +1711,14 @@ test = ["pytest", "pytest-xdist", "setuptools"]
 
 [[package]]
 name = "pygments"
-version = "2.19.1"
+version = "2.19.2"
 description = "Pygments is a syntax highlighting package written in Python."
 optional = false
 python-versions = ">=3.8"
 groups = ["main"]
 files = [
-    {file = "pygments-2.19.1-py3-none-any.whl", hash = "sha256:9ea1544ad55cecf4b8242fab6dd35a93bbce657034b0611ee383099054ab6d8c"},
-    {file = "pygments-2.19.1.tar.gz", hash = "sha256:61c16d2a8576dc0649d9f39e089b5f02bcd27fba10d8fb4dcc28173f7a45151f"},
+    {file = "pygments-2.19.2-py3-none-any.whl", hash = "sha256:86540386c03d588bb81d44bc3928634ff26449851e99741617ecb9037ee5ec0b"},
+    {file = "pygments-2.19.2.tar.gz", hash = "sha256:636cb2477cec7f8952536970bc533bc43743542f70392ae026374600add5b887"},
 ]
 
 [package.extras]
@@ -1720,14 +1726,14 @@ windows-terminal = ["colorama (>=0.4.6)"]
 
 [[package]]
 name = "pymdown-extensions"
-version = "10.15"
+version = "10.16"
 description = "Extension pack for Python Markdown."
 optional = false
-python-versions = ">=3.8"
+python-versions = ">=3.9"
 groups = ["main"]
 files = [
-    {file = "pymdown_extensions-10.15-py3-none-any.whl", hash = "sha256:46e99bb272612b0de3b7e7caf6da8dd5f4ca5212c0b273feb9304e236c484e5f"},
-    {file = "pymdown_extensions-10.15.tar.gz", hash = "sha256:0e5994e32155f4b03504f939e501b981d306daf7ec2aa1cd2eb6bd300784f8f7"},
+    {file = "pymdown_extensions-10.16-py3-none-any.whl", hash = "sha256:f5dd064a4db588cb2d95229fc4ee63a1b16cc8b4d0e6145c0899ed8723da1df2"},
+    {file = "pymdown_extensions-10.16.tar.gz", hash = "sha256:71dac4fca63fabeffd3eb9038b756161a33ec6e8d230853d3cecf562155ab3de"},
 ]
 
 [package.dependencies]
diff --git a/pyproject.toml b/pyproject.toml
index 3052ad3..c66e62e 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -16,7 +16,7 @@ dependencies = [
     "mkdocs (>=1.6.1,<2.0.0)",
     "mkdocstrings[python] (>=0.29.1,<0.30.0)",
     "mkdocs-material (>=9.6.11,<10.0.0)",
-    "scipy (>=1.15.2,<2.0.0)"
+    "scipy (>=1.15.2,<2.0.0)",
 ]
 
 [tool.pytest.ini_options]
diff --git a/robot_nav/SIM_ENV/__init__.py b/robot_nav/SIM_ENV/__init__.py
new file mode 100644
index 0000000..e69de29
diff --git a/robot_nav/SIM_ENV/marl_sim.py b/robot_nav/SIM_ENV/marl_sim.py
new file mode 100644
index 0000000..8fd10ba
--- /dev/null
+++ b/robot_nav/SIM_ENV/marl_sim.py
@@ -0,0 +1,300 @@
+import irsim
+import numpy as np
+import random
+import torch
+
+from robot_nav.SIM_ENV.sim_env import SIM_ENV
+
+
+class MARL_SIM(SIM_ENV):
+    """
+    A simulation environment interface for robot navigation using IRSim in MARL setting.
+
+    This class wraps around the IRSim environment and provides methods for stepping,
+    resetting, and interacting with mobile robots, including reward computation.
+
+    Attributes:
+        env (object): The simulation environment instance from IRSim.
+        robot_goal (np.ndarray): The goal position of the robot.
+    """
+
+    def __init__(self, world_file="multi_robot_world.yaml", disable_plotting=False):
+        """
+        Initialize the simulation environment.
+
+        Args:
+            world_file (str): Path to the world configuration YAML file.
+            disable_plotting (bool): If True, disables rendering and plotting.
+        """
+        display = False if disable_plotting else True
+        self.env = irsim.make(
+            world_file, disable_all_plot=disable_plotting, display=display
+        )
+        robot_info = self.env.get_robot_info(0)
+        self.robot_goal = robot_info.goal
+        self.num_robots = len(self.env.robot_list)
+        self.x_range = self.env._world.x_range
+        self.y_range = self.env._world.y_range
+
+    def step(self, action, connection, combined_weights=None):
+        """
+        Perform one step in the simulation using the given control commands.
+
+        Args:
+            lin_velocity (float): Linear velocity to apply to the robot.
+            ang_velocity (float): Angular velocity to apply to the robot.
+
+        Returns:
+            (tuple): Contains the latest LIDAR scan, distance to goal, cosine and sine of angle to goal,
+                   collision flag, goal reached flag, applied action, and computed reward.
+        """
+        self.env.step(action_id=[i for i in range(self.num_robots)], action=action)
+        self.env.render()
+
+        poses = []
+        distances = []
+        coss = []
+        sins = []
+        collisions = []
+        goals = []
+        rewards = []
+        positions = []
+        goal_positions = []
+        robot_states = [
+            [self.env.robot_list[i].state[0], self.env.robot_list[i].state[1]]
+            for i in range(self.num_robots)
+        ]
+        for i in range(self.num_robots):
+
+            robot_state = self.env.robot_list[i].state
+            closest_robots = [
+                np.linalg.norm(
+                    [
+                        robot_states[j][0] - robot_state[0],
+                        robot_states[j][1] - robot_state[1],
+                    ]
+                )
+                for j in range(self.num_robots)
+                if j != i
+            ]
+            robot_goal = self.env.robot_list[i].goal
+            goal_vector = [
+                robot_goal[0].item() - robot_state[0].item(),
+                robot_goal[1].item() - robot_state[1].item(),
+            ]
+            distance = np.linalg.norm(goal_vector)
+            goal = self.env.robot_list[i].arrive
+            pose_vector = [np.cos(robot_state[2]).item(), np.sin(robot_state[2]).item()]
+            cos, sin = self.cossin(pose_vector, goal_vector)
+            collision = self.env.robot_list[i].collision
+            action_i = action[i]
+            reward = self.get_reward(
+                goal, collision, action_i, closest_robots, distance
+            )
+
+            position = [robot_state[0].item(), robot_state[1].item()]
+            goal_position = [robot_goal[0].item(), robot_goal[1].item()]
+
+            distances.append(distance)
+            coss.append(cos)
+            sins.append(sin)
+            collisions.append(collision)
+            goals.append(goal)
+            rewards.append(reward)
+            positions.append(position)
+            poses.append(
+                [robot_state[0].item(), robot_state[1].item(), robot_state[2].item()]
+            )
+            goal_positions.append(goal_position)
+
+            i_probs = torch.sigmoid(
+                connection[i]
+            )  # connection[i] is logits for "connect" per pair
+
+            # Now we need to insert the self-connection (optional, typically skipped)
+            i_connections = i_probs.tolist()
+            i_connections.insert(i, 0)
+            if combined_weights is not None:
+                i_weights = combined_weights[i].tolist()
+                i_weights.insert(i, 0)
+
+            for j in range(self.num_robots):
+                if i_connections[j] > 0.5:
+                    if combined_weights is not None:
+                        weight = i_weights[j]
+                    else:
+                        weight = 1
+                    other_robot_state = self.env.robot_list[j].state
+                    other_pos = [
+                        other_robot_state[0].item(),
+                        other_robot_state[1].item(),
+                    ]
+                    rx = [position[0], other_pos[0]]
+                    ry = [position[1], other_pos[1]]
+                    self.env.draw_trajectory(
+                        np.array([rx, ry]), refresh=True, linewidth=weight
+                    )
+
+            if goal:
+                self.env.robot_list[i].set_random_goal(
+                    obstacle_list=self.env.obstacle_list,
+                    init=True,
+                    range_limits=[
+                        [self.x_range[0] + 1, self.y_range[0] + 1, -3.141592653589793],
+                        [self.x_range[1] - 1, self.y_range[1] - 1, 3.141592653589793],
+                    ],
+                )
+
+        return (
+            poses,
+            distances,
+            coss,
+            sins,
+            collisions,
+            goals,
+            action,
+            rewards,
+            positions,
+            goal_positions,
+        )
+
+    def reset(
+        self,
+        robot_state=None,
+        robot_goal=None,
+        random_obstacles=False,
+        random_obstacle_ids=None,
+    ):
+        """
+        Reset the simulation environment, optionally setting robot and obstacle states.
+
+        Args:
+            robot_state (list or None): Initial state of the robot as a list of [x, y, theta, speed].
+            robot_goal (list or None): Goal state for the robot.
+            random_obstacles (bool): Whether to randomly reposition obstacles.
+            random_obstacle_ids (list or None): Specific obstacle IDs to randomize.
+
+        Returns:
+            (tuple): Initial observation after reset, including LIDAR scan, distance, cos/sin,
+                   and reward-related flags and values.
+        """
+        if robot_state is None:
+            robot_state = [[random.uniform(3, 9)], [random.uniform(3, 9)], [0]]
+
+        init_states = []
+        for robot in self.env.robot_list:
+            conflict = True
+            while conflict:
+                conflict = False
+                robot_state = [
+                    [random.uniform(3, 9)],
+                    [random.uniform(3, 9)],
+                    [random.uniform(-3.14, 3.14)],
+                ]
+                pos = [robot_state[0][0], robot_state[1][0]]
+                for loc in init_states:
+                    vector = [
+                        pos[0] - loc[0],
+                        pos[1] - loc[1],
+                    ]
+                    if np.linalg.norm(vector) < 0.6:
+                        conflict = True
+            init_states.append(pos)
+
+            robot.set_state(
+                state=np.array(robot_state),
+                init=True,
+            )
+
+        if random_obstacles:
+            if random_obstacle_ids is None:
+                random_obstacle_ids = [i + self.num_robots for i in range(7)]
+            self.env.random_obstacle_position(
+                range_low=[self.x_range[0], self.y_range[0], -3.14],
+                range_high=[self.x_range[1], self.y_range[1], 3.14],
+                ids=random_obstacle_ids,
+                non_overlapping=True,
+            )
+
+        for robot in self.env.robot_list:
+            if robot_goal is None:
+                robot.set_random_goal(
+                    obstacle_list=self.env.obstacle_list,
+                    init=True,
+                    range_limits=[
+                        [self.x_range[0] + 1, self.y_range[0] + 1, -3.141592653589793],
+                        [self.x_range[1] - 1, self.y_range[1] - 1, 3.141592653589793],
+                    ],
+                )
+            else:
+                self.env.robot.set_goal(np.array(robot_goal), init=True)
+        self.env.reset()
+        self.robot_goal = self.env.robot.goal
+
+        action = [[0.0, 0.0] for _ in range(self.num_robots)]
+        con = torch.tensor(
+            [[0.0 for _ in range(self.num_robots - 1)] for _ in range(self.num_robots)]
+        )
+        poses, distance, cos, sin, _, _, action, reward, positions, goal_positions = (
+            self.step(action, con)
+        )
+        return (
+            poses,
+            distance,
+            cos,
+            sin,
+            [False] * self.num_robots,
+            [False] * self.num_robots,
+            action,
+            reward,
+            positions,
+            goal_positions,
+        )
+
+    @staticmethod
+    def get_reward(goal, collision, action, closest_robots, distance, phase=1):
+        """
+        Calculate the reward for the current step.
+
+        Args:
+            goal (bool): Whether the goal has been reached.
+            collision (bool): Whether a collision occurred.
+            action (list): The action taken [linear velocity, angular velocity].
+            closest_robots (list): Distances to the closest robots.
+            distance (float): Distance to goal.
+            phase (int, optional): Reward function phase. Defaults to 1.
+
+        Returns:
+            (float): Computed reward for the current state.
+        """
+
+        match phase:
+            case 1:
+                if goal:
+                    return 100.0
+                elif collision:
+                    return -100.0 * 3 * action[0]
+                else:
+                    r_dist = 1.5 / distance
+                    cl_pen = 0
+                    for rob in closest_robots:
+                        add = 1.5 - rob if rob < 1.5 else 0
+                        cl_pen += add
+
+                    return action[0] - 0.5 * abs(action[1]) - cl_pen + r_dist
+
+            case 2:
+                if goal:
+                    return 70.0
+                elif collision:
+                    return -100.0 * 3 * action[0]
+                else:
+                    cl_pen = 0
+                    for rob in closest_robots:
+                        add = (3 - rob) ** 2 if rob < 3 else 0
+                        cl_pen += add
+
+                    return -0.5 * abs(action[1]) - cl_pen
+
+            case _:
+                raise ValueError("Unknown reward phase")
diff --git a/robot_nav/sim.py b/robot_nav/SIM_ENV/sim.py
similarity index 89%
rename from robot_nav/sim.py
rename to robot_nav/SIM_ENV/sim.py
index adfe7f0..c341373 100644
--- a/robot_nav/sim.py
+++ b/robot_nav/SIM_ENV/sim.py
@@ -2,11 +2,10 @@
 import numpy as np
 import random
 
-import shapely
-from irsim.lib.handler.geometry_handler import GeometryFactory
+from robot_nav.SIM_ENV.sim_env import SIM_ENV
 
 
-class SIM_ENV:
+class SIM(SIM_ENV):
     """
     A simulation environment interface for robot navigation using IRSim.
 
@@ -121,24 +120,6 @@ def reset(
         )
         return latest_scan, distance, cos, sin, False, False, action, reward
 
-    @staticmethod
-    def cossin(vec1, vec2):
-        """
-        Compute the cosine and sine of the angle between two 2D vectors.
-
-        Args:
-            vec1 (list): First 2D vector.
-            vec2 (list): Second 2D vector.
-
-        Returns:
-            (tuple): (cosine, sine) of the angle between the vectors.
-        """
-        vec1 = vec1 / np.linalg.norm(vec1)
-        vec2 = vec2 / np.linalg.norm(vec2)
-        cos = np.dot(vec1, vec2)
-        sin = vec1[0] * vec2[1] - vec1[1] * vec2[0]
-        return cos, sin
-
     @staticmethod
     def get_reward(goal, collision, action, laser_scan):
         """
diff --git a/robot_nav/SIM_ENV/sim_env.py b/robot_nav/SIM_ENV/sim_env.py
new file mode 100644
index 0000000..4cd6f77
--- /dev/null
+++ b/robot_nav/SIM_ENV/sim_env.py
@@ -0,0 +1,35 @@
+from abc import ABC, abstractmethod
+import numpy as np
+
+
+class SIM_ENV(ABC):
+    @abstractmethod
+    def step(self):
+        raise NotImplementedError("step method must be implemented by subclass.")
+
+    @abstractmethod
+    def reset(self):
+        raise NotImplementedError("reset method must be implemented by subclass.")
+
+    @staticmethod
+    def cossin(vec1, vec2):
+        """
+        Compute the cosine and sine of the angle between two 2D vectors.
+
+        Args:
+            vec1 (list): First 2D vector.
+            vec2 (list): Second 2D vector.
+
+        Returns:
+            (tuple): (cosine, sine) of the angle between the vectors.
+        """
+        vec1 = vec1 / np.linalg.norm(vec1)
+        vec2 = vec2 / np.linalg.norm(vec2)
+        cos = np.dot(vec1, vec2)
+        sin = vec1[0] * vec2[1] - vec1[1] * vec2[0]
+        return cos, sin
+
+    @staticmethod
+    @abstractmethod
+    def get_reward():
+        raise NotImplementedError("get_reward method must be implemented by subclass.")
diff --git a/robot_nav/models/CNNTD3/CNNTD3.py b/robot_nav/models/CNNTD3/CNNTD3.py
index 0be20da..af6cf45 100644
--- a/robot_nav/models/CNNTD3/CNNTD3.py
+++ b/robot_nav/models/CNNTD3/CNNTD3.py
@@ -325,8 +325,8 @@ def train(
             state = torch.Tensor(batch_states).to(self.device)
             next_state = torch.Tensor(batch_next_states).to(self.device)
             action = torch.Tensor(batch_actions).to(self.device)
-            reward = torch.Tensor(batch_rewards).to(self.device)
-            done = torch.Tensor(batch_dones).to(self.device)
+            reward = torch.Tensor(batch_rewards).to(self.device).reshape(-1, 1)
+            done = torch.Tensor(batch_dones).to(self.device).reshape(-1, 1)
 
             # Obtain the estimated action from the next state by using the actor-target
             next_action = self.actor_target(next_state)
diff --git a/robot_nav/models/DDPG/DDPG.py b/robot_nav/models/DDPG/DDPG.py
index 0f9b98b..2b3abfe 100644
--- a/robot_nav/models/DDPG/DDPG.py
+++ b/robot_nav/models/DDPG/DDPG.py
@@ -254,8 +254,8 @@ def train(
             state = torch.Tensor(batch_states).to(self.device)
             next_state = torch.Tensor(batch_next_states).to(self.device)
             action = torch.Tensor(batch_actions).to(self.device)
-            reward = torch.Tensor(batch_rewards).to(self.device)
-            done = torch.Tensor(batch_dones).to(self.device)
+            reward = torch.Tensor(batch_rewards).to(self.device).reshape(-1, 1)
+            done = torch.Tensor(batch_dones).to(self.device).reshape(-1, 1)
 
             # Obtain the estimated action from the next state by using the actor-target
             next_action = self.actor_target(next_state)
diff --git a/robot_nav/models/MARL/hardsoftAttention.py b/robot_nav/models/MARL/hardsoftAttention.py
new file mode 100644
index 0000000..4978a9f
--- /dev/null
+++ b/robot_nav/models/MARL/hardsoftAttention.py
@@ -0,0 +1,222 @@
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+
+class Attention(nn.Module):
+    def __init__(self, embedding_dim):
+        super(Attention, self).__init__()
+        self.embedding_dim = embedding_dim
+
+        self.embedding1 = nn.Linear(5, 128)
+        nn.init.kaiming_uniform_(self.embedding1.weight, nonlinearity="leaky_relu")
+        self.embedding2 = nn.Linear(128, embedding_dim)
+        nn.init.kaiming_uniform_(self.embedding2.weight, nonlinearity="leaky_relu")
+
+        # Hard attention MLP with distance
+        self.hard_mlp = nn.Sequential(
+            nn.Linear(embedding_dim + 7, embedding_dim),
+            nn.ReLU(),
+            nn.Linear(embedding_dim, embedding_dim),
+        )
+        self.hard_encoding = nn.Linear(embedding_dim, 2)
+
+        # Soft attention projections
+        self.q = nn.Linear(embedding_dim, embedding_dim, bias=False)
+        self.k = nn.Linear(10, embedding_dim, bias=False)
+        self.v = nn.Linear(10, embedding_dim)
+
+        # Soft attention score network (with polar other robot goal position)
+        self.attn_score_layer = nn.Sequential(
+            nn.Linear(embedding_dim * 2, embedding_dim),
+            nn.ReLU(),
+            nn.Linear(embedding_dim, 1),
+        )
+
+        # Decoder
+        self.decode_1 = nn.Linear(embedding_dim * 2, embedding_dim * 2)
+        nn.init.kaiming_uniform_(self.decode_1.weight, nonlinearity="leaky_relu")
+        self.decode_2 = nn.Linear(embedding_dim * 2, embedding_dim * 2)
+        nn.init.kaiming_uniform_(self.decode_2.weight, nonlinearity="leaky_relu")
+
+    def encode_agent_features(self, embed):
+        embed = F.leaky_relu(self.embedding1(embed))
+        embed = F.leaky_relu(self.embedding2(embed))
+        return embed
+
+    def forward(self, embedding):
+        if embedding.dim() == 2:
+            embedding = embedding.unsqueeze(0)
+        batch_size, n_agents, _ = embedding.shape
+
+        embed = embedding[:, :, 4:9].reshape(batch_size * n_agents, -1)
+        position = embedding[:, :, :2].reshape(batch_size, n_agents, 2)
+        heading = embedding[:, :, 2:4].reshape(
+            batch_size, n_agents, 2
+        )  # assume (cos(θ), sin(θ))
+        action = embedding[:, :, 7:9].reshape(batch_size, n_agents, 2)
+        goal = embedding[:, :, -2:].reshape(batch_size, n_agents, 2)
+        goal_j = goal.unsqueeze(1).expand(-1, n_agents, -1, -1)
+        pos_i = position.unsqueeze(2)
+        goal_rel_vec = goal_j - pos_i
+
+        agent_embed = self.encode_agent_features(embed)
+        agent_embed = agent_embed.view(batch_size, n_agents, self.embedding_dim)
+
+        # For hard attention
+        h_i = agent_embed.unsqueeze(2)  # (B, N, 1, D)
+        pos_i = position.unsqueeze(2)  # (B, N, 1, 2)
+        pos_j = position.unsqueeze(1)  # (B, 1, N, 2)
+        heading_i = heading.unsqueeze(2)  # (B, N, 1, 2)
+        heading_j = heading.unsqueeze(1).expand(-1, n_agents, -1, -1)  # (B, 1, N, 2)
+
+        # Compute relative vectors and distance
+        rel_vec = pos_j - pos_i  # (B, N, N, 2)
+        dx, dy = rel_vec[..., 0], rel_vec[..., 1]
+        rel_dist = (
+            torch.linalg.vector_norm(rel_vec, dim=-1, keepdim=True) / 12
+        )  # (B, N, N, 1)
+
+        # Relative angle in agent i's frame
+        angle = torch.atan2(dy, dx) - torch.atan2(heading_i[..., 1], heading_i[..., 0])
+        angle = (angle + np.pi) % (2 * np.pi) - np.pi
+        rel_angle_sin = torch.sin(angle)
+        rel_angle_cos = torch.cos(angle)
+
+        # Other agent's heading
+        heading_j_cos = heading_j[..., 0]  # (B, 1, N)
+        heading_j_sin = heading_j[..., 1]  # (B, 1, N)
+
+        # Stack edge features
+        edge_features = torch.cat(
+            [
+                rel_dist,  # (B, N, N, 1)
+                rel_angle_cos.unsqueeze(-1),  # (B, N, N, 1)
+                rel_angle_sin.unsqueeze(-1),  # (B, N, N, 1)
+                heading_j_cos.unsqueeze(-1),  # (B, N, N, 1)
+                heading_j_sin.unsqueeze(-1),  # (B, N, N, 1)
+                action.unsqueeze(1).expand(-1, n_agents, -1, -1),  # (B, N, N, 2)
+            ],
+            dim=-1,
+        )
+
+        # Broadcast h_i along N (for each pair)
+        h_i_expanded = h_i.expand(-1, -1, n_agents, -1)
+
+        # Remove self-pairs using mask
+        mask = ~torch.eye(n_agents, dtype=torch.bool, device=embedding.device)
+        h_i_flat = h_i_expanded[:, mask].reshape(
+            batch_size * n_agents, n_agents - 1, self.embedding_dim
+        )
+        edge_flat = edge_features[:, mask].reshape(
+            batch_size * n_agents, n_agents - 1, -1
+        )
+
+        # Concatenate agent embedding and edge features
+        hard_input = torch.cat([h_i_flat, edge_flat], dim=-1)
+
+        # Hard attention forward
+        h_hard = self.hard_mlp(hard_input)
+        hard_logits = self.hard_encoding(h_hard)
+        hard_weights = F.gumbel_softmax(hard_logits, hard=False, tau=0.5, dim=-1)[
+            ..., 1
+        ].unsqueeze(2)
+        hard_weights = hard_weights.view(batch_size, n_agents, n_agents - 1)
+
+        unnorm_rel_dist = torch.linalg.vector_norm(rel_vec, dim=-1, keepdim=True)
+        unnorm_rel_dist = unnorm_rel_dist[:, mask].reshape(
+            batch_size * n_agents, n_agents - 1, 1
+        )
+
+        # Soft attention
+        q = self.q(agent_embed)
+
+        attention_outputs = []
+        entropy_list = []
+        combined_w = []
+
+        goal_rel_dist = torch.linalg.vector_norm(goal_rel_vec, dim=-1, keepdim=True)
+        goal_angle_global = torch.atan2(goal_rel_vec[..., 1], goal_rel_vec[..., 0])
+        heading_angle = torch.atan2(heading_i[..., 1], heading_i[..., 0])
+        goal_rel_angle = goal_angle_global - heading_angle
+        goal_rel_angle = (goal_rel_angle + np.pi) % (2 * np.pi) - np.pi
+        goal_rel_angle_cos = torch.cos(goal_rel_angle).unsqueeze(-1)
+        goal_rel_angle_sin = torch.sin(goal_rel_angle).unsqueeze(-1)
+        goal_polar = torch.cat(
+            [goal_rel_dist, goal_rel_angle_cos, goal_rel_angle_sin], dim=-1
+        )
+
+        soft_edge_features = torch.cat([edge_features, goal_polar], dim=-1)
+        for i in range(n_agents):
+            q_i = q[:, i : i + 1, :]
+            mask = torch.ones(n_agents, dtype=torch.bool, device=edge_features.device)
+            mask[i] = False
+            edge_i_wo_self = soft_edge_features[:, i, mask, :]
+            edge_i_wo_self = edge_i_wo_self.squeeze(1)
+            k = F.leaky_relu(self.k(edge_i_wo_self))
+
+            q_i_expanded = q_i.expand(-1, n_agents - 1, -1)
+            attention_input = torch.cat([q_i_expanded, k], dim=-1)
+
+            # Score computation
+            scores = self.attn_score_layer(attention_input).transpose(1, 2)
+
+            # Mask using hard weights
+            h_weights = hard_weights[:, i].unsqueeze(1)
+            mask = (h_weights > 0.5).float()
+
+            # All-zero mask handling
+            epsilon = 1e-6
+            mask_sum = mask.sum(dim=-1, keepdim=True)
+            all_zero_mask = mask_sum < epsilon
+
+            # Apply mask to scores
+            masked_scores = scores.masked_fill(mask == 0, float("-inf"))
+
+            # Compute softmax, safely handle all-zero cases
+            soft_weights = F.softmax(masked_scores, dim=-1)
+            soft_weights = torch.where(
+                all_zero_mask, torch.zeros_like(soft_weights), soft_weights
+            )
+
+            combined_weights = soft_weights * mask  # (B, 1, N-1)
+            combined_w.append(combined_weights)
+
+            # Normalize combined_weights for entropy calculation
+            combined_weights_norm = combined_weights / (
+                combined_weights.sum(dim=-1, keepdim=True) + epsilon
+            )
+
+            # Calculate entropy from combined_weights
+            entropy = (
+                -(combined_weights_norm * (combined_weights_norm + epsilon).log())
+                .sum(dim=-1)
+                .mean()
+            )
+            entropy_list.append(entropy)
+
+            v_j = F.leaky_relu(self.v(edge_i_wo_self))
+            attn_output = torch.bmm(combined_weights, v_j).squeeze(1)
+            attention_outputs.append(attn_output)
+
+        comb_w = torch.stack(combined_w, dim=1).reshape(n_agents, -1)
+        attn_stack = torch.stack(attention_outputs, dim=1).reshape(
+            -1, self.embedding_dim
+        )
+        self_embed = agent_embed.reshape(-1, self.embedding_dim)
+        concat_embed = torch.cat([self_embed, attn_stack], dim=-1)
+
+        x = F.leaky_relu(self.decode_1(concat_embed))
+        att_embedding = F.leaky_relu(self.decode_2(x))
+
+        mean_entropy = torch.stack(entropy_list).mean()
+
+        return (
+            att_embedding,
+            hard_logits[..., 1],
+            unnorm_rel_dist,
+            mean_entropy,
+            hard_weights,
+            comb_w,
+        )
diff --git a/robot_nav/models/MARL/marlTD3.py b/robot_nav/models/MARL/marlTD3.py
new file mode 100644
index 0000000..9fd4508
--- /dev/null
+++ b/robot_nav/models/MARL/marlTD3.py
@@ -0,0 +1,506 @@
+from pathlib import Path
+
+import numpy as np
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+from numpy import inf
+from torch.utils.tensorboard import SummaryWriter
+
+from robot_nav.models.MARL.hardsoftAttention import Attention
+
+
+class Actor(nn.Module):
+    def __init__(self, action_dim, embedding_dim):
+        super().__init__()
+        self.attention = Attention(embedding_dim)  # ➊ edge classifier
+
+        # ➋ policy head (everything _after_ attention)
+        self.policy_head = nn.Sequential(
+            nn.Linear(embedding_dim * 2, 400),
+            nn.LeakyReLU(),
+            nn.Linear(400, 300),
+            nn.LeakyReLU(),
+            nn.Linear(300, action_dim),
+            nn.Tanh(),
+        )
+
+    def forward(self, obs, detach_attn=False):
+        attn_out, hard_logits, pair_d, mean_entropy, hard_weights, combined_weights = (
+            self.attention(obs)
+        )
+        if detach_attn:  # used in the policy phase
+            attn_out = attn_out.detach()
+        action = self.policy_head(attn_out)
+        return action, hard_logits, pair_d, mean_entropy, hard_weights, combined_weights
+
+
+class Critic(nn.Module):
+    def __init__(self, action_dim, embedding_dim):
+        super(Critic, self).__init__()
+        self.embedding_dim = embedding_dim
+        self.attention = Attention(self.embedding_dim)
+
+        self.layer_1 = nn.Linear(self.embedding_dim * 2, 400)
+        torch.nn.init.kaiming_uniform_(self.layer_1.weight, nonlinearity="leaky_relu")
+
+        self.layer_2_s = nn.Linear(400, 300)
+        torch.nn.init.kaiming_uniform_(self.layer_2_s.weight, nonlinearity="leaky_relu")
+
+        self.layer_2_a = nn.Linear(action_dim, 300)
+        torch.nn.init.kaiming_uniform_(self.layer_2_a.weight, nonlinearity="leaky_relu")
+
+        self.layer_3 = nn.Linear(300, 1)
+        torch.nn.init.kaiming_uniform_(self.layer_3.weight, nonlinearity="leaky_relu")
+
+        self.layer_4 = nn.Linear(self.embedding_dim * 2, 400)
+        torch.nn.init.kaiming_uniform_(
+            self.layer_4.weight, nonlinearity="leaky_relu"
+        )  # ✅ Fixed init bug
+
+        self.layer_5_s = nn.Linear(400, 300)
+        torch.nn.init.kaiming_uniform_(self.layer_5_s.weight, nonlinearity="leaky_relu")
+
+        self.layer_5_a = nn.Linear(action_dim, 300)
+        torch.nn.init.kaiming_uniform_(self.layer_5_a.weight, nonlinearity="leaky_relu")
+
+        self.layer_6 = nn.Linear(300, 1)
+        torch.nn.init.kaiming_uniform_(self.layer_6.weight, nonlinearity="leaky_relu")
+
+    def forward(self, embedding, action):
+
+        (
+            embedding_with_attention,
+            hard_logits,
+            unnorm_rel_dist,
+            mean_entropy,
+            hard_weights,
+            _,
+        ) = self.attention(embedding)
+
+        # Q1
+        s1 = F.leaky_relu(self.layer_1(embedding_with_attention))
+        s1 = F.leaky_relu(self.layer_2_s(s1) + self.layer_2_a(action))  # ✅ No .data
+        q1 = self.layer_3(s1)
+
+        # Q2
+        s2 = F.leaky_relu(self.layer_4(embedding_with_attention))
+        s2 = F.leaky_relu(self.layer_5_s(s2) + self.layer_5_a(action))  # ✅ No .data
+        q2 = self.layer_6(s2)
+
+        return q1, q2, mean_entropy, hard_logits, unnorm_rel_dist, hard_weights
+
+
+class TD3(object):
+    """
+    CNNTD3 (Twin Delayed Deep Deterministic Policy Gradient with CNN-based inputs) agent for
+    continuous control tasks.
+
+    This class encapsulates the full implementation of the TD3 algorithm using neural network
+    architectures for the actor and critic, with optional bounding for critic outputs to
+    regularize learning. The agent is designed to train in environments where sensor
+    observations (e.g., LiDAR) are used for navigation tasks.
+
+    Args:
+        state_dim (int): Dimension of the input state.
+        action_dim (int): Dimension of the output action.
+        max_action (float): Maximum magnitude of the action.
+        device (torch.device): Torch device to use (CPU or GPU).
+        lr (float): Learning rate for both actor and critic optimizers.
+        save_every (int): Save model every N training iterations (0 to disable).
+        load_model (bool): Whether to load a pre-trained model at initialization.
+        save_directory (Path): Path to the directory for saving model checkpoints.
+        model_name (str): Base name for the saved model files.
+        load_directory (Path): Path to load model checkpoints from (if `load_model=True`).
+        use_max_bound (bool): Whether to apply maximum Q-value bounding during training.
+        bound_weight (float): Weight for the bounding loss term in total loss.
+    """
+
+    def __init__(
+        self,
+        state_dim,
+        action_dim,
+        max_action,
+        device,
+        num_robots,
+        lr_actor=1e-4,
+        lr_critic=3e-4,
+        save_every=0,
+        load_model=False,
+        save_directory=Path("robot_nav/models/MARL/checkpoint"),
+        model_name="marlTD3",
+        load_model_name=None,
+        load_directory=Path("robot_nav/models/MARL/checkpoint"),
+    ):
+        # Initialize the Actor network
+        self.num_robots = num_robots
+        self.device = device
+        self.actor = Actor(action_dim, embedding_dim=256).to(
+            self.device
+        )  # Using the updated Actor
+        self.actor_target = Actor(action_dim, embedding_dim=256).to(self.device)
+        self.actor_target.load_state_dict(self.actor.state_dict())
+
+        self.attn_params = list(self.actor.attention.parameters())
+        self.policy_params = list(self.actor.policy_head.parameters())
+
+        self.actor_optimizer = torch.optim.Adam(
+            self.policy_params + self.attn_params, lr=lr_actor
+        )  # TD3 policy
+
+        self.critic = Critic(action_dim, embedding_dim=256).to(
+            self.device
+        )  # Using the updated Critic
+        self.critic_target = Critic(action_dim, embedding_dim=256).to(self.device)
+        self.critic_target.load_state_dict(self.critic.state_dict())
+        self.critic_optimizer = torch.optim.Adam(
+            params=self.critic.parameters(), lr=lr_critic
+        )
+        self.action_dim = action_dim
+        self.max_action = max_action
+        self.state_dim = state_dim
+        self.writer = SummaryWriter(comment=model_name)
+        self.iter_count = 0
+        if load_model_name is None:
+            load_model_name = model_name
+        if load_model:
+            self.load(filename=load_model_name, directory=load_directory)
+        self.save_every = save_every
+        self.model_name = model_name
+        self.save_directory = save_directory
+
+    def get_action(self, obs, add_noise):
+        """
+        Selects an action for a given observation.
+
+        Args:
+            obs (np.ndarray): The current observation/state.
+            add_noise (bool): Whether to add exploration noise to the action.
+
+        Returns:
+            (np.ndarray): The selected action.
+        """
+        action, connection, combined_weights = self.act(obs)
+        if add_noise:
+            noise = np.random.normal(0, 0.5, size=action.shape)
+            noise = [n / 4 if i % 2 else n for i, n in enumerate(noise)]
+            action = (action + noise).clip(-self.max_action, self.max_action)
+
+        return action.reshape(-1, 2), connection, combined_weights
+
+    def act(self, state):
+        """
+        Computes the deterministic action from the actor network for a given state.
+
+        Args:
+            state (np.ndarray): Input state.
+
+        Returns:
+            (np.ndarray): Action predicted by the actor network.
+        """
+        # Function to get the action from the actor
+        state = torch.Tensor(state).to(self.device)
+        # res = self.attention(state)
+        action, connection, _, _, _, combined_weights = self.actor(state)
+        return action.cpu().data.numpy().flatten(), connection, combined_weights
+
+    # training cycle
+    def train(
+        self,
+        replay_buffer,
+        iterations,
+        batch_size,
+        discount=0.99,
+        tau=0.005,
+        policy_noise=0.2,
+        noise_clip=0.5,
+        policy_freq=2,
+        bce_weight=0.1,
+        entropy_weight=1,
+        connection_proximity_threshold=4,
+    ):
+        av_Q = 0
+        max_Q = -inf
+        av_loss = 0
+        av_critic_loss = 0
+        av_critic_entropy = []
+        av_actor_entropy = []
+        av_actor_loss = 0
+        av_critic_bce_loss = []
+        av_actor_bce_loss = []
+
+        for it in range(iterations):
+            # sample a batch
+            (
+                batch_states,
+                batch_actions,
+                batch_rewards,
+                batch_dones,
+                batch_next_states,
+            ) = replay_buffer.sample_batch(batch_size)
+
+            state = (
+                torch.Tensor(batch_states)
+                .to(self.device)
+                .view(batch_size, self.num_robots, self.state_dim)
+            )
+            next_state = (
+                torch.Tensor(batch_next_states)
+                .to(self.device)
+                .view(batch_size, self.num_robots, self.state_dim)
+            )
+            action = (
+                torch.Tensor(batch_actions)
+                .to(self.device)
+                .view(batch_size * self.num_robots, self.action_dim)
+            )
+            reward = (
+                torch.Tensor(batch_rewards)
+                .to(self.device)
+                .view(batch_size * self.num_robots, 1)
+            )
+            done = (
+                torch.Tensor(batch_dones)
+                .to(self.device)
+                .view(batch_size * self.num_robots, 1)
+            )
+
+            with torch.no_grad():
+                next_action, _, _, _, _, _ = self.actor_target(
+                    next_state, detach_attn=True
+                )
+
+            # --- Target smoothing ---
+            noise = (
+                torch.Tensor(batch_actions)
+                .data.normal_(0, policy_noise)
+                .to(self.device)
+            ).reshape(-1, 2)
+            noise = noise.clamp(-noise_clip, noise_clip)
+            next_action = (next_action + noise).clamp(-self.max_action, self.max_action)
+
+            # --- Target Q values ---
+            target_Q1, target_Q2, _, _, _, _ = self.critic_target(
+                next_state, next_action
+            )
+            target_Q = torch.min(target_Q1, target_Q2)
+            av_Q += target_Q.mean()
+            max_Q = max(max_Q, target_Q.max().item())
+            target_Q = reward + ((1 - done) * discount * target_Q).detach()
+
+            # --- Critic update ---
+            (
+                current_Q1,
+                current_Q2,
+                mean_entropy,
+                hard_logits,
+                unnorm_rel_dist,
+                hard_weights,
+            ) = self.critic(state, action)
+            critic_loss = F.mse_loss(current_Q1, target_Q) + F.mse_loss(
+                current_Q2, target_Q
+            )
+
+            targets = (
+                unnorm_rel_dist.flatten() < connection_proximity_threshold
+            ).float()
+            flat_logits = hard_logits.flatten()
+            bce_loss = F.binary_cross_entropy_with_logits(flat_logits, targets)
+
+            av_critic_bce_loss.append(bce_loss)
+
+            total_loss = (
+                critic_loss - entropy_weight * mean_entropy + bce_weight * bce_loss
+            )
+            av_critic_entropy.append(mean_entropy)
+
+            self.critic_optimizer.zero_grad()
+            total_loss.backward()
+            torch.nn.utils.clip_grad_norm_(self.critic.parameters(), 10.0)
+            self.critic_optimizer.step()
+
+            av_loss += total_loss.item()
+            av_critic_loss += critic_loss.item()
+
+            # --- Actor update ---
+            if it % policy_freq == 0:
+
+                action, hard_logits, unnorm_rel_dist, mean_entropy, hard_weights, _ = (
+                    self.actor(state, detach_attn=False)
+                )
+                targets = (
+                    unnorm_rel_dist.flatten() < connection_proximity_threshold
+                ).float()
+                flat_logits = hard_logits.flatten()
+                bce_loss = F.binary_cross_entropy_with_logits(flat_logits, targets)
+
+                av_actor_bce_loss.append(bce_loss)
+
+                actor_Q, _, _, _, _, _ = self.critic(state, action)
+                actor_loss = -actor_Q.mean()
+                total_loss = (
+                    actor_loss - entropy_weight * mean_entropy + bce_weight * bce_loss
+                )
+                av_actor_entropy.append(mean_entropy)
+
+                self.actor_optimizer.zero_grad()
+                total_loss.backward()
+                torch.nn.utils.clip_grad_norm_(self.policy_params, 10.0)
+                self.actor_optimizer.step()
+
+                av_actor_loss += total_loss.item()
+
+                # Soft update target networks
+                for param, target_param in zip(
+                    self.actor.parameters(), self.actor_target.parameters()
+                ):
+                    target_param.data.copy_(
+                        tau * param.data + (1 - tau) * target_param.data
+                    )
+
+                for param, target_param in zip(
+                    self.critic.parameters(), self.critic_target.parameters()
+                ):
+                    target_param.data.copy_(
+                        tau * param.data + (1 - tau) * target_param.data
+                    )
+
+        self.iter_count += 1
+        self.writer.add_scalar(
+            "train/loss_total", av_loss / iterations, self.iter_count
+        )
+        self.writer.add_scalar(
+            "train/critic_loss", av_critic_loss / iterations, self.iter_count
+        )
+        self.writer.add_scalar(
+            "train/av_critic_entropy",
+            sum(av_critic_entropy) / len(av_critic_entropy),
+            self.iter_count,
+        )
+        self.writer.add_scalar(
+            "train/av_actor_entropy",
+            sum(av_actor_entropy) / len(av_actor_entropy),
+            self.iter_count,
+        )
+        self.writer.add_scalar(
+            "train/av_critic_bce_loss",
+            sum(av_critic_bce_loss) / len(av_critic_bce_loss),
+            self.iter_count,
+        )
+        self.writer.add_scalar(
+            "train/av_actor_bce_loss",
+            sum(av_actor_bce_loss) / len(av_actor_bce_loss),
+            self.iter_count,
+        )
+        self.writer.add_scalar("train/avg_Q", av_Q / iterations, self.iter_count)
+        self.writer.add_scalar("train/max_Q", max_Q, self.iter_count)
+
+        self.writer.add_scalar(
+            "train/actor_loss",
+            av_actor_loss / (iterations // policy_freq),
+            self.iter_count,
+        )
+
+        if self.save_every > 0 and self.iter_count % self.save_every == 0:
+            self.save(filename=self.model_name, directory=self.save_directory)
+
+    def save(self, filename, directory):
+        """
+        Saves the current model parameters to the specified directory.
+
+        Args:
+            filename (str): Base filename for saved files.
+            directory (Path): Path to save the model files.
+        """
+        Path(directory).mkdir(parents=True, exist_ok=True)
+        torch.save(self.actor.state_dict(), "%s/%s_actor.pth" % (directory, filename))
+        torch.save(
+            self.actor_target.state_dict(),
+            "%s/%s_actor_target.pth" % (directory, filename),
+        )
+        torch.save(self.critic.state_dict(), "%s/%s_critic.pth" % (directory, filename))
+        torch.save(
+            self.critic_target.state_dict(),
+            "%s/%s_critic_target.pth" % (directory, filename),
+        )
+
+    def load(self, filename, directory):
+        """
+        Loads model parameters from the specified directory.
+
+        Args:
+            filename (str): Base filename for saved files.
+            directory (Path): Path to load the model files from.
+        """
+        self.actor.load_state_dict(
+            torch.load("%s/%s_actor.pth" % (directory, filename))
+        )
+        self.actor_target.load_state_dict(
+            torch.load("%s/%s_actor_target.pth" % (directory, filename))
+        )
+        self.critic.load_state_dict(
+            torch.load("%s/%s_critic.pth" % (directory, filename))
+        )
+        self.critic_target.load_state_dict(
+            torch.load("%s/%s_critic_target.pth" % (directory, filename))
+        )
+        print(f"Loaded weights from: {directory}")
+
+    def prepare_state(
+        self, poses, distance, cos, sin, collision, action, goal_positions
+    ):
+        """
+        Prepares the environment's raw agent state for learning.
+
+        Args:
+            poses (list): Each agent's global pose [x, y, theta].
+            distance, cos, sin: Unused, can be removed or ignored.
+            collision (list): Collision flags per agent.
+            action (list): Last action taken [lin_vel, ang_vel].
+            goal_positions (list): Each agent's goal [x, y].
+
+        Returns:
+            states (list): List of processed state vectors.
+            terminal (list): Flags (1 if collision or goal reached, else 0).
+        """
+        states = []
+        terminal = []
+
+        for i in range(self.num_robots):
+            pose = poses[i]  # [x, y, theta]
+            goal_pos = goal_positions[i]  # [goal_x, goal_y]
+            act = action[i]  # [lin_vel, ang_vel]
+
+            px, py, theta = pose
+            gx, gy = goal_pos
+
+            # Heading as cos/sin
+            heading_cos = np.cos(theta)
+            heading_sin = np.sin(theta)
+
+            # Last velocity
+            lin_vel = act[0] * 2  # Assuming original range [0, 0.5]
+            ang_vel = (act[1] + 1) / 2  # Assuming original range [-1, 1]
+
+            # Final state vector
+            state = [
+                px,
+                py,
+                heading_cos,
+                heading_sin,
+                distance[i] / 17,
+                cos[i],
+                sin[i],
+                lin_vel,
+                ang_vel,
+                gx,
+                gy,
+            ]
+
+            assert (
+                len(state) == self.state_dim
+            ), f"State length mismatch: expected {self.state_dim}, got {len(state)}"
+            states.append(state)
+            terminal.append(collision[i])
+
+        return states, terminal
diff --git a/robot_nav/models/SAC/SAC.py b/robot_nav/models/SAC/SAC.py
index 10639ae..157e1d3 100644
--- a/robot_nav/models/SAC/SAC.py
+++ b/robot_nav/models/SAC/SAC.py
@@ -346,8 +346,8 @@ def update(self, replay_buffer, step, batch_size):
         state = torch.Tensor(batch_states).to(self.device)
         next_state = torch.Tensor(batch_next_states).to(self.device)
         action = torch.Tensor(batch_actions).to(self.device)
-        reward = torch.Tensor(batch_rewards).to(self.device)
-        done = torch.Tensor(batch_dones).to(self.device)
+        reward = torch.Tensor(batch_rewards).to(self.device).reshape(-1, 1)
+        done = torch.Tensor(batch_dones).to(self.device).reshape(-1, 1)
         self.train_metrics_dict["train/batch_reward_av"].append(
             batch_rewards.mean().item()
         )
diff --git a/robot_nav/models/TD3/TD3.py b/robot_nav/models/TD3/TD3.py
index c35943b..8c66666 100644
--- a/robot_nav/models/TD3/TD3.py
+++ b/robot_nav/models/TD3/TD3.py
@@ -268,8 +268,8 @@ def train(
             state = torch.Tensor(batch_states).to(self.device)
             next_state = torch.Tensor(batch_next_states).to(self.device)
             action = torch.Tensor(batch_actions).to(self.device)
-            reward = torch.Tensor(batch_rewards).to(self.device)
-            done = torch.Tensor(batch_dones).to(self.device)
+            reward = torch.Tensor(batch_rewards).to(self.device).reshape(-1, 1)
+            done = torch.Tensor(batch_dones).to(self.device).reshape(-1, 1)
 
             # Obtain the estimated action from the next state by using the actor-target
             next_action = self.actor_target(next_state)
@@ -445,7 +445,7 @@ def prepare_state(self, latest_scan, distance, cos, sin, collision, goal, action
         ang_vel = (action[1] + 1) / 2
         state = min_values + [distance, cos, sin] + [lin_vel, ang_vel]
 
-        assert len(state) == self.state_dim
+        assert len(state) == self.state_dim, f"{len(state), self.state_dim}"
         terminal = 1 if collision or goal else 0
 
         return state, terminal
diff --git a/robot_nav/multi_robot_world.yaml b/robot_nav/multi_robot_world.yaml
new file mode 100644
index 0000000..66e31ec
--- /dev/null
+++ b/robot_nav/multi_robot_world.yaml
@@ -0,0 +1,22 @@
+world:
+  height: 12  # the height of the world
+  width: 12   # the height of the world
+  step_time: 0.3  # Calculate each step
+  sample_time: 0.3  # For render and data extraction
+  collision_mode: 'reactive'
+
+robot:
+  - number: 5
+    kinematics: {name: 'diff'}
+    distribution: {name: 'circle', radius: 4, center: [6, 6]}
+    shape: {name: 'circle', radius: 0.2}
+    vel_min: [ 0, -1.0 ]
+    vel_max: [ 1.0, 1.0 ]
+    state: [2, 2, 0, 0]
+    goal: [9, 9, 0]
+    color: ['royalblue', 'red', 'green', 'orange', 'purple', 'yellow', 'cyan', 'magenta', 'lime', 'pink', 'brown']
+    arrive_mode: position
+    goal_threshold: 0.3
+
+    plot:
+      show_trajectory: False
diff --git a/robot_nav/replay_buffer.py b/robot_nav/replay_buffer.py
index 8c8eeea..f5902d7 100644
--- a/robot_nav/replay_buffer.py
+++ b/robot_nav/replay_buffer.py
@@ -71,8 +71,8 @@ def sample_batch(self, batch_size):
 
         s_batch = np.array([_[0] for _ in batch])
         a_batch = np.array([_[1] for _ in batch])
-        r_batch = np.array([_[2] for _ in batch]).reshape(-1, 1)
-        t_batch = np.array([_[3] for _ in batch]).reshape(-1, 1)
+        r_batch = np.array([_[2] for _ in batch])
+        t_batch = np.array([_[3] for _ in batch])
         s2_batch = np.array([_[4] for _ in batch])
 
         return s_batch, a_batch, r_batch, t_batch, s2_batch
diff --git a/robot_nav/test.py b/robot_nav/test.py
index fe59051..087f570 100644
--- a/robot_nav/test.py
+++ b/robot_nav/test.py
@@ -1,13 +1,8 @@
 from models.TD3.TD3 import TD3
-from models.DDPG.DDPG import DDPG
-from models.SAC.SAC import SAC
-from models.HCM.hardcoded_model import HCM
-from models.PPO.PPO import PPO
-from robot_nav.models.CNNTD3.CNNTD3 import CNNTD3
 
 import torch
 import numpy as np
-from sim import SIM_ENV
+from robot_nav.SIM_ENV.sim import SIM
 import yaml
 
 
@@ -31,7 +26,7 @@ def main(args=None):
         model_name="TD3",
     )  # instantiate a model
 
-    sim = SIM_ENV(world_file="eval_world.yaml")  # instantiate environment
+    sim = SIM(world_file="eval_world.yaml")  # instantiate environment
     with open("robot_nav/eval_points.yaml") as file:
         points = yaml.safe_load(file)
     robot_poses = points["robot"]["poses"]
diff --git a/robot_nav/test_random.py b/robot_nav/test_random.py
index 90446a2..63690ec 100644
--- a/robot_nav/test_random.py
+++ b/robot_nav/test_random.py
@@ -1,16 +1,11 @@
-from robot_nav.models.TD3.TD3 import TD3
-from robot_nav.models.DDPG.DDPG import DDPG
 from robot_nav.models.SAC.SAC import SAC
-from robot_nav.models.HCM.hardcoded_model import HCM
-from robot_nav.models.PPO.PPO import PPO
-from robot_nav.models.CNNTD3.CNNTD3 import CNNTD3
 import statistics
 import numpy as np
 import tqdm
 import matplotlib.pyplot as plt
 
 import torch
-from sim import SIM_ENV
+from robot_nav.SIM_ENV.sim import SIM
 
 
 def main(args=None):
@@ -34,7 +29,7 @@ def main(args=None):
         model_name="SAC",
     )  # instantiate a model
 
-    sim = SIM_ENV(
+    sim = SIM(
         world_file="eval_world.yaml", disable_plotting=False
     )  # instantiate environment
 
diff --git a/robot_nav/test_rnn.py b/robot_nav/test_rnn.py
index 526ee34..9fca0b9 100644
--- a/robot_nav/test_rnn.py
+++ b/robot_nav/test_rnn.py
@@ -4,7 +4,7 @@
 
 import torch
 import numpy as np
-from sim import SIM_ENV
+from robot_nav.SIM_ENV.sim import SIM
 import yaml
 
 
@@ -27,7 +27,7 @@ def main(args=None):
         load_model=True,
     )  # instantiate a model
 
-    sim = SIM_ENV(world_file="eval_world.yaml")  # instantiate environment
+    sim = SIM(world_file="eval_world.yaml")  # instantiate environment
     with open("robot_nav/eval_points.yaml") as file:
         points = yaml.safe_load(file)
     robot_poses = points["robot"]["poses"]
diff --git a/robot_nav/train.py b/robot_nav/train.py
index 936ecc7..6d97362 100644
--- a/robot_nav/train.py
+++ b/robot_nav/train.py
@@ -1,13 +1,8 @@
-from robot_nav.models.TD3.TD3 import TD3
-from robot_nav.models.DDPG.DDPG import DDPG
-from robot_nav.models.SAC.SAC import SAC
-from robot_nav.models.HCM.hardcoded_model import HCM
-from robot_nav.models.PPO.PPO import PPO
 from robot_nav.models.CNNTD3.CNNTD3 import CNNTD3
 
 import torch
 import numpy as np
-from sim import SIM_ENV
+from robot_nav.SIM_ENV.sim import SIM
 from utils import get_buffer
 
 
@@ -15,7 +10,7 @@ def main(args=None):
     """Main training function"""
     action_dim = 2  # number of actions produced by the model
     max_action = 1  # maximum absolute value of output actions
-    state_dim = 25  # number of input values in the neural network (vector length of state input)
+    state_dim = 95  # number of input values in the neural network (vector length of state input)
     device = torch.device(
         "cuda" if torch.cuda.is_available() else "cpu"
     )  # using cuda if it is available, cpu otherwise
@@ -36,17 +31,19 @@ def main(args=None):
     )
     save_every = 5  # save the model every n training cycles
 
-    model = TD3(
+    model = CNNTD3(
         state_dim=state_dim,
         action_dim=action_dim,
         max_action=max_action,
         device=device,
         save_every=save_every,
         load_model=False,
-        model_name="TD3",
+        model_name="CNNTD3",
     )  # instantiate a model
 
-    sim = SIM_ENV(disable_plotting=False)  # instantiate environment
+    sim = SIM(
+        world_file="robot_world.yaml", disable_plotting=False
+    )  # instantiate environment
     replay_buffer = get_buffer(
         model,
         sim,
diff --git a/robot_nav/train_marl.py b/robot_nav/train_marl.py
new file mode 100644
index 0000000..52e4dfa
--- /dev/null
+++ b/robot_nav/train_marl.py
@@ -0,0 +1,154 @@
+from robot_nav.models.MARL.marlTD3 import TD3
+
+import torch
+import numpy as np
+from robot_nav.SIM_ENV.marl_sim import MARL_SIM
+from utils import get_buffer
+
+
+def outside_of_bounds(poses):
+    outside = False
+    for pose in poses:
+        norm_x = pose[0] - 6
+        norm_y = pose[1] - 6
+        if abs(norm_x) > 10.5 or abs(norm_y) > 10.5:
+            outside = True
+            break
+    return outside
+
+
+def main(args=None):
+    """Main training function"""
+    action_dim = 2  # number of actions produced by the model
+    max_action = 1  # maximum absolute value of output actions
+    state_dim = 11  # number of input values in the neural network (vector length of state input)
+    device = torch.device(
+        "cuda" if torch.cuda.is_available() else "cpu"
+    )  # using cuda if it is available, cpu otherwise
+    max_epochs = 600  # max number of epochs
+    epoch = 1  # starting epoch number
+    episode = 0  # starting episode number
+    train_every_n = 10  # train and update network parameters every n episodes
+    training_iterations = 80  # how many batches to use for single training cycle
+    batch_size = 16  # batch size for each training iteration
+    max_steps = 300  # maximum number of steps in single episode
+    steps = 0  # starting step number
+    load_saved_buffer = False  # whether to load experiences from assets/data.yml
+    pretrain = False  # whether to use the loaded experiences to pre-train the model (load_saved_buffer must be True)
+    pretraining_iterations = (
+        10  # number of training iterations to run during pre-training
+    )
+    save_every = 5  # save the model every n training cycles
+
+    sim = MARL_SIM(
+        world_file="multi_robot_world.yaml", disable_plotting=False
+    )  # instantiate environment
+
+    model = TD3(
+        state_dim=state_dim,
+        action_dim=action_dim,
+        max_action=max_action,
+        num_robots=sim.num_robots,
+        device=device,
+        save_every=save_every,
+        load_model=False,
+        model_name="phase1",
+        load_model_name="phase1",
+    )  # instantiate a model
+
+    replay_buffer = get_buffer(
+        model,
+        sim,
+        load_saved_buffer,
+        pretrain,
+        pretraining_iterations,
+        training_iterations,
+        batch_size,
+    )
+    connections = torch.tensor(
+        [[0.0 for _ in range(sim.num_robots - 1)] for _ in range(sim.num_robots)]
+    )
+
+    poses, distance, cos, sin, collision, goal, a, reward, positions, goal_positions = (
+        sim.step([[0, 0] for _ in range(sim.num_robots)], connections)
+    )  # get the initial step state
+    running_goals = 0
+    running_collisions = 0
+    running_timesteps = 0
+    while epoch < max_epochs:  # train until max_epochs is reached
+        state, terminal = model.prepare_state(
+            poses, distance, cos, sin, collision, a, goal_positions
+        )  # get state a state representation from returned data from the environment
+
+        action, connection, combined_weights = model.get_action(
+            np.array(state), True
+        )  # get an action from the model
+
+        a_in = [
+            [(a[0] + 1) / 4, a[1]] for a in action
+        ]  # clip linear velocity to [0, 0.5] m/s range
+
+        (
+            poses,
+            distance,
+            cos,
+            sin,
+            collision,
+            goal,
+            a,
+            reward,
+            positions,
+            goal_positions,
+        ) = sim.step(
+            a_in, connection, combined_weights
+        )  # get data from the environment
+        running_goals += sum(goal)
+        running_collisions += sum(collision)
+        running_timesteps += 1
+        next_state, terminal = model.prepare_state(
+            poses, distance, cos, sin, collision, a, goal_positions
+        )  # get a next state representation
+        replay_buffer.add(
+            state, action, reward, terminal, next_state
+        )  # add experience to the replay buffer
+        outside = outside_of_bounds(poses)
+        if (
+            any(terminal) or steps == max_steps or outside
+        ):  # reset environment of terminal state reached, or max_steps were taken
+            (
+                poses,
+                distance,
+                cos,
+                sin,
+                collision,
+                goal,
+                a,
+                reward,
+                positions,
+                goal_positions,
+            ) = sim.reset()
+            episode += 1
+            if episode % train_every_n == 0:
+                model.writer.add_scalar(
+                    "run/avg_goal", running_goals / running_timesteps, epoch
+                )
+                model.writer.add_scalar(
+                    "run/avg_collision", running_collisions / running_timesteps, epoch
+                )
+                running_goals = 0
+                running_collisions = 0
+                running_timesteps = 0
+                epoch += 1
+                model.train(
+                    replay_buffer=replay_buffer,
+                    iterations=training_iterations,
+                    batch_size=batch_size,
+                )  # train the model and update its parameters
+
+            steps = 0
+        else:
+            steps += 1
+
+
+if __name__ == "__main__":
+    main()
diff --git a/robot_nav/train_rnn.py b/robot_nav/train_rnn.py
index aec1f70..5d1ad4e 100644
--- a/robot_nav/train_rnn.py
+++ b/robot_nav/train_rnn.py
@@ -5,7 +5,7 @@
 import numpy as np
 
 
-from sim import SIM_ENV
+from robot_nav.SIM_ENV.sim import SIM
 from utils import get_buffer
 
 
@@ -47,7 +47,7 @@ def main(args=None):
         rnn="gru",
     )  # instantiate a model
 
-    sim = SIM_ENV()  # instantiate environment
+    sim = SIM()  # instantiate environment
     replay_buffer = get_buffer(
         model,
         sim,
diff --git a/tests/test_marl_world.yaml b/tests/test_marl_world.yaml
new file mode 100644
index 0000000..6af7f68
--- /dev/null
+++ b/tests/test_marl_world.yaml
@@ -0,0 +1,22 @@
+world:
+  height: 12  # the height of the world
+  width: 12   # the height of the world
+  step_time: 1.0
+  sample_time: 1.0
+  collision_mode: 'reactive'
+
+robot:
+  - number: 3
+    kinematics: {name: 'diff'}
+    distribution: {name: 'manual'}
+    shape: {name: 'circle', radius: 0.2}
+    vel_min: [ 0, -1.0 ]
+    vel_max: [ 1.0, 1.0 ]
+    state: [[3, 10, 0], [3, 6, 0], [3, 2, 0]]
+    goal: [[9, 9, 0], [8, 8, 0], [7, 7, 0]]
+    color: ['royalblue', 'red', 'green', 'orange', 'purple', 'yellow', 'cyan', 'magenta', 'lime', 'pink', 'brown']
+    arrive_mode: position
+    goal_threshold: 0.3
+
+    plot:
+      show_trajectory: False
diff --git a/tests/test_model.py b/tests/test_model.py
index 13193ac..6ad13d7 100644
--- a/tests/test_model.py
+++ b/tests/test_model.py
@@ -1,16 +1,21 @@
 from pathlib import Path
 
+import torch
+
+from robot_nav.SIM_ENV.marl_sim import MARL_SIM
 from robot_nav.models.RCPG.RCPG import RCPG
 from robot_nav.models.TD3.TD3 import TD3
 from robot_nav.models.CNNTD3.CNNTD3 import CNNTD3
 from robot_nav.models.SAC.SAC import SAC
 from robot_nav.models.DDPG.DDPG import DDPG
 from robot_nav.utils import get_buffer
-from robot_nav.sim import SIM_ENV
+from robot_nav.SIM_ENV.sim import SIM
+from robot_nav.models.MARL.marlTD3 import TD3 as marlTD3
 import pytest
 
 PROJECT_ROOT = Path(__file__).resolve().parents[1]
 
+
 @pytest.mark.parametrize(
     "model, state_dim",
     [
@@ -31,7 +36,7 @@ def test_models(model, state_dim):
         load_model=False,
     )  # instantiate a model
 
-    sim = SIM_ENV
+    sim = SIM
 
     prefilled_buffer = get_buffer(
         model=test_model,
@@ -71,7 +76,7 @@ def test_max_bound_models(model, state_dim):
         use_max_bound=True,
     )  # instantiate a model
 
-    sim = SIM_ENV
+    sim = SIM
 
     prefilled_buffer = get_buffer(
         model=test_model,
@@ -90,3 +95,56 @@ def test_max_bound_models(model, state_dim):
         iterations=2,
         batch_size=8,
     )
+
+
+def test_marl_models():
+    sim = MARL_SIM("/tests/test_marl_world.yaml", disable_plotting=True)
+    test_model = marlTD3(
+        state_dim=11,
+        action_dim=2,
+        max_action=1,
+        num_robots=sim.num_robots,
+        device="cpu",
+        save_every=0,
+        load_model=False,
+    )  # instantiate a model
+
+    replay_buffer = get_buffer(
+        model=test_model,
+        sim=sim,
+        load_saved_buffer=False,
+        pretrain=False,
+        pretraining_iterations=0,
+        training_iterations=0,
+        batch_size=0,
+        buffer_size=100,
+    )
+
+    for _ in range(10):
+        connections = torch.tensor(
+            [[0.0 for _ in range(sim.num_robots - 1)] for _ in range(3)]
+        )
+        (
+            poses,
+            distance,
+            cos,
+            sin,
+            collision,
+            goal,
+            a,
+            reward,
+            positions,
+            goal_positions,
+        ) = sim.step([[0, 0] for _ in range(sim.num_robots)], connections)
+        state, terminal = test_model.prepare_state(
+            poses, distance, cos, sin, collision, a, goal_positions
+        )
+        replay_buffer.add(
+            state, [[0, 0] for _ in range(sim.num_robots)], reward, terminal, state
+        )
+
+    test_model.train(
+        replay_buffer=replay_buffer,
+        iterations=2,
+        batch_size=8,
+    )
diff --git a/tests/test_sim.py b/tests/test_sim.py
index adb9e01..cf2e616 100644
--- a/tests/test_sim.py
+++ b/tests/test_sim.py
@@ -1,18 +1,20 @@
 import os
 
 import pytest
+import torch
 
-from robot_nav.sim import SIM_ENV
+from robot_nav.SIM_ENV.marl_sim import MARL_SIM
+from robot_nav.SIM_ENV.sim import SIM
 import numpy as np
 
 skip_on_ci = pytest.mark.skipif(
-    os.getenv("CI") == "true",
-    reason="Skipped on CI (GitHub Actions)"
+    os.getenv("CI") == "true", reason="Skipped on CI (GitHub Actions)"
 )
 
+
 @skip_on_ci
 def test_sim():
-    sim = SIM_ENV("/tests/test_world.yaml")
+    sim = SIM("/tests/test_world.yaml", disable_plotting=True)
     robot_state = sim.env.get_robot_state()
     state = sim.step(1, 0)
     next_robot_state = sim.env.get_robot_state()
@@ -27,9 +29,32 @@ def test_sim():
     assert np.not_equal(robot_state[0], new_robot_state[0])
     assert np.not_equal(robot_state[1], new_robot_state[1])
 
+
+def test_marl_sim():
+    sim = MARL_SIM("/tests/test_marl_world.yaml", disable_plotting=True)
+    robot_state = [sim.env.robot_list[i].state[:2] for i in range(3)]
+    connections = torch.tensor(
+        [[0.0 for _ in range(sim.num_robots - 1)] for _ in range(3)]
+    )
+
+    _ = sim.step([[1, 0], [1, 0], [1, 0]], connections)
+    next_robot_state = [sim.env.robot_list[i].state[:2] for i in range(3)]
+    for j in range(3):
+        assert np.isclose(robot_state[j][0], next_robot_state[j][0] - 1)
+        assert np.isclose(robot_state[j][1], robot_state[j][1])
+
+    assert len(sim.env.obstacle_list) == 0
+
+    sim.reset()
+    new_robot_state = [sim.env.robot_list[i].state[:2] for i in range(3)]
+    for j in range(3):
+        assert np.not_equal(robot_state[j][0], new_robot_state[j][0])
+        assert np.not_equal(robot_state[j][1], new_robot_state[j][1])
+
+
 @skip_on_ci
 def test_sincos():
-    sim = SIM_ENV("/tests/test_world.yaml")
+    sim = SIM("/tests/test_world.yaml")
     cos, sin = sim.cossin([1, 0], [0, 1])
     assert np.isclose(cos, 0)
     assert np.isclose(sin, 1)
diff --git a/tests/test_utils.py b/tests/test_utils.py
index 7aa18ef..2414152 100644
--- a/tests/test_utils.py
+++ b/tests/test_utils.py
@@ -4,10 +4,11 @@
 from robot_nav.models.PPO.PPO import PPO, RolloutBuffer
 from robot_nav.models.RCPG.RCPG import RCPG
 from robot_nav.utils import get_buffer, RolloutReplayBuffer, ReplayBuffer
-from robot_nav.sim import SIM_ENV
+from robot_nav.SIM_ENV.sim import SIM
 
 PROJECT_ROOT = Path(__file__).resolve().parents[1]
 
+
 def test_buffer():
     model = SAC(
         state_dim=10,
@@ -18,7 +19,7 @@ def test_buffer():
         load_model=False,
     )  # instantiate a model
 
-    sim = SIM_ENV
+    sim = SIM
     buffer = get_buffer(
         model=model,
         sim=sim,
@@ -57,7 +58,7 @@ def test_rollout_buffer():
         load_model=False,
     )  # instantiate a model
 
-    sim = SIM_ENV
+    sim = SIM
     buffer = get_buffer(
         model=model,
         sim=sim,
@@ -95,7 +96,7 @@ def test_ppo_buffer():
         save_every=0,
         load_model=False,
     )
-    sim = SIM_ENV
+    sim = SIM
     buffer = get_buffer(
         model=model,
         sim=sim,