Need help in a quadroped simulation #2692
Replies: 1 comment
-
Hi, I am a MuJoCo user. I hope this helps you if you haven't figured it out already! The behaviour mentioned by you is expected. The robot's weight gets distributed amongst the supporting legs based on the location of the robot's centre of mass. Visualising the robot's COM could help. You would see a large tilt if the COM projection is close to the edges of the support polygon (formed by leg touch points) and the legs are not stiff enough. You could do some trials like: use higher PD gains (with the same ratio between kp and kv) and try moving up only one leg at a time You could get more help if you shared the complete XML model along with the mesh files and the code. |
Beta Was this translation helpful? Give feedback.
Uh oh!
There was an error while loading. Please reload this page.
-
Intro
Hi!
I am a undergraduate student at IITR , I use MuJoCo for my research on quadropeds.
My setup
Mujoco, Python, win11
My question
Issue Description:
When I simulate my quadruped with the torso fixed in the air, the leg trajectories of the forelegs and hind legs are identical, and the motion appears symmetrical as expected. However, once I release the torso and allow the robot to interact with the ground (i.e., in a free simulation), the hind legs do not lift off the ground as much as the forelegs.
Despite identical trajectory commands being sent to all legs, the hind legs remain closer to the ground throughout the gait cycle. This tilts forward slightly, keeping the hind legs in near-constant contact with the ground.
Minimal model and/or code that explain my question
Confirmations
Beta Was this translation helpful? Give feedback.
All reactions