Merge pull request #7 from NVIDIA-ISAAC-ROS/release-dp3.1

jaiveersinghNV · web-flow · commit 7de0fcc684c4 · 2023-05-25T19:51:44.000-07:00
Isaac ROS 0.31.0 (DP3.1)
diff --git a/README.md b/README.md
@@ -35,9 +35,10 @@ This package is powered by [NVIDIA Isaac Transport for ROS (NITROS)](https://dev
 
 The following table summarizes the per-platform performance statistics of sample graphs that use this package, with links included to the full benchmark output. These benchmark configurations are taken from the [Isaac ROS Benchmark](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark#list-of-isaac-ros-benchmarks) collection, based on the [`ros2_benchmark`](https://github.com/NVIDIA-ISAAC-ROS/ros2_benchmark) framework.
 
-| Sample Graph                                                                                                                                         | Input Size | AGX Orin                                                                                                                                           | Orin NX                                                                                                                                            | Orin Nano 8GB                                                                                                                                            | x86_64 w/ RTX 3060 Ti                                                                                                                                       |
-| ---------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | -------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| [Occupancy Grid Localizer Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/scripts//isaac_ros_grid_localizer_node.py) | ~50 sq. m         | [16.0 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/results/isaac_ros_grid_localizer_node-agx_orin.json)<br>79 ms | [7.33 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/results/isaac_ros_grid_localizer_node-orin_nx.json)<br>150 ms | [5.16 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/results/isaac_ros_grid_localizer_node-orin_nano_8gb.json)<br>220 ms | [46.5 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/results/isaac_ros_grid_localizer_node-x86_64_rtx_3060Ti.json)<br>33 ms |
+| Sample Graph                                                                                                                                         | Input Size | AGX Orin                                                                                                                                           | Orin NX                                                                                                                                            | Orin Nano 8GB                                                                                                                                        | x86_64 w/ RTX 4060 Ti                                                                                                                                |
+| ---------------------------------------------------------------------------------------------------------------------------------------------------- | ---------- | -------------------------------------------------------------------------------------------------------------------------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------- |
+| [Occupancy Grid Localizer Node](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/scripts//isaac_ros_grid_localizer_node.py) | ~50 sq. m  | [15.9 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/results/isaac_ros_grid_localizer_node-agx_orin.json)<br>65 ms | [7.33 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/results/isaac_ros_grid_localizer_node-orin_nx.json)<br>140 ms | [5.17 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/results/isaac_ros_grid_localizer_node-orin_nano.json)<br>200 ms | [50.1 fps](https://github.com/NVIDIA-ISAAC-ROS/isaac_ros_benchmark/blob/main/results/isaac_ros_grid_localizer_node-nuc_4060ti.json)<br>20 ms |
+
 
 
 ## Table of Contents
@@ -81,7 +82,7 @@ The following table summarizes the per-platform performance statistics of sample
 
 ## Latest Update
 
-2023-04-05: Added `isaac_ros_occupancy_grid_localizer`
+2023-05-25: Performance improvements.
 
 ## Supported Platforms
 
@@ -362,4 +363,5 @@ For solutions to problems with Isaac ROS, please check [here](https://github.com
 
 | Date       | Changes                                    |
 | ---------- | ------------------------------------------ |
+| 2023-05-25 | Performance improvements                   |
 | 2023-04-05 | Added `isaac_ros_occupancy_grid_localizer` |
diff --git a/isaac_ros_occupancy_grid_localizer/config/occupancy_grid_localizer_node.yaml b/isaac_ros_occupancy_grid_localizer/config/occupancy_grid_localizer_node.yaml
@@ -221,7 +221,7 @@ components:
     allocator: allocator
     cuda_stream: cuda_stream/cuda_stream
 ---
-name: vault
+name: sink
 components:
 - name: signal
   type: nvidia::gxf::DoubleBufferReceiver
@@ -232,12 +232,10 @@ components:
   parameters:
     receiver: signal
     min_size: 1
-- name: vault
-  type: nvidia::gxf::Vault
+- name: sink
+  type: nvidia::isaac_ros::MessageRelay
   parameters:
     source: signal
-    max_waiting_count: 1
-    drop_waiting: false
 ---
 components:
 - name: edge0
@@ -264,16 +262,16 @@ components:
   type: nvidia::gxf::Connection
   parameters:
     source: localization_to_pose3d/tx_pose3d_cov
-    target: vault/signal
+    target: sink/signal
 ---
 components:
 - name: clock
   type: nvidia::gxf::RealtimeClock
-- type: nvidia::gxf::MultiThreadScheduler
+- type: nvidia::gxf::GreedyScheduler
   parameters:
     clock: clock
-    worker_thread_number: 8
     stop_on_deadlock: false
+    check_recession_period_us: 100
 - type: nvidia::gxf::JobStatistics
   parameters:
     clock: clock
diff --git a/isaac_ros_occupancy_grid_localizer/package.xml b/isaac_ros_occupancy_grid_localizer/package.xml
@@ -21,7 +21,7 @@
 <?xml-model href="http://download.ros.org/schema/package_format3.xsd" schematypens="http://www.w3.org/2001/XMLSchema"?>
 <package format="3">
   <name>isaac_ros_occupancy_grid_localizer</name>
-  <version>0.30.0</version>
+  <version>0.31.0</version>
   <description>Occupancy Grid Global Localizer.</description>
 
   <maintainer email="ashwinvk@nvidia.com">Ashwin Varghese Kuruttukulam</maintainer>
diff --git a/isaac_ros_occupancy_grid_localizer/src/occupancy_grid_localizer_node.cpp b/isaac_ros_occupancy_grid_localizer/src/occupancy_grid_localizer_node.cpp
@@ -74,7 +74,7 @@ constexpr char INPUT_COMPONENT_KEY_FLATSCAN[] = "flatscan_beams_tensor_copier/rx
 constexpr char INPUT_DEFAULT_TENSOR_FORMAT_FLATSCAN[] = "nitros_flat_scan";
 constexpr char INPUT_TOPIC_NAME_FLATSCAN[] = "flatscan_localization";
 
-constexpr char OUTPUT_COMPONENT_KEY_LOC_RESULT[] = "vault/vault";
+constexpr char OUTPUT_COMPONENT_KEY_LOC_RESULT[] = "sink/sink";
 constexpr char OUTPUT_DEFAULT_TENSOR_FORMAT_LOC_RESULT[] = "nitros_pose_cov_stamped";
 constexpr char OUTPUT_TOPIC_NAME_LOC_RESULT[] = "localization_result";
 constexpr char OUTPUT_FRAME_ID_MAP_KEY[] = "map";
diff --git a/isaac_ros_pointcloud_utils/config/pointcloud_to_flatscan_node.yaml b/isaac_ros_pointcloud_utils/config/pointcloud_to_flatscan_node.yaml
@@ -56,9 +56,9 @@ components:
     max_points: 691200
 ---
 ###########################
-# Vault for ROS Publisher #
+# Sink for ROS Publisher #
 ###########################
-name: vault
+name: sink
 components:
 - name: signal
   type: nvidia::gxf::DoubleBufferReceiver
@@ -69,28 +69,26 @@ components:
   parameters:
     receiver: signal
     min_size: 1
-- name: vault
-  type: nvidia::gxf::Vault
+- name: sink
+  type: nvidia::isaac_ros::MessageRelay
   parameters:
     source: signal
-    max_waiting_count: 1
-    drop_waiting: false
 ---
 components:
 - name: edge0
   type: nvidia::gxf::Connection
   parameters:
     source: pointcloud_to_flatscan/tx_flatscan
-    target: vault/signal
+    target: sink/signal
 ---
 components:
 - name: clock
   type: nvidia::gxf::RealtimeClock
-- type: nvidia::gxf::MultiThreadScheduler
+- type: nvidia::gxf::GreedyScheduler
   parameters:
     clock: clock
-    worker_thread_number: 8
     stop_on_deadlock: false
+    check_recession_period_us: 100
 - type: nvidia::gxf::JobStatistics
   parameters:
     clock: clock
diff --git a/isaac_ros_pointcloud_utils/package.xml b/isaac_ros_pointcloud_utils/package.xml
@@ -21,7 +21,7 @@
 <?xml-model href="http://download.ros.org/schema/package_format3.xsd" schematypens="http://www.w3.org/2001/XMLSchema"?>
 <package format="3">
   <name>isaac_ros_pointcloud_utils</name>
-  <version>0.30.0</version>
+  <version>0.31.0</version>
   <description>Isaac ROS Point Cloud Utilities.</description>
 
   <maintainer email="ashwinvk@nvidia.com">Ashwin Varghese Kuruttukulam</maintainer>
diff --git a/isaac_ros_pointcloud_utils/src/pointcloud_to_flatscan_node.cpp b/isaac_ros_pointcloud_utils/src/pointcloud_to_flatscan_node.cpp
@@ -54,7 +54,7 @@ constexpr char INPUT_COMPONENT_KEY_POINTCLOUD[] = "pointcloud_to_flatscan/rx_poi
 constexpr char INPUT_DEFAULT_TENSOR_FORMAT_POINTCLOUD[] = "nitros_point_cloud";
 constexpr char INPUT_TOPIC_NAME_POINTCLOUD[] = "pointcloud";
 
-constexpr char OUTPUT_COMPONENT_KEY_FLATSCAN[] = "vault/vault";
+constexpr char OUTPUT_COMPONENT_KEY_FLATSCAN[] = "sink/sink";
 constexpr char OUTPUT_DEFAULT_TENSOR_FORMAT_FLATSCAN[] = "nitros_flat_scan";
 constexpr char OUTPUT_TOPIC_NAME_FLATSCAN[] = "flatscan";
 
