search

chvmp / champ

MIT Cheetah I Implementation
BSD 3-Clause "New" or "Revised" License
1.57k stars 361 forks source link

Unitree A1 robot fails to move in Gazebo and falls over #32

Closed tahir1069 closed 3 years ago

tahir1069 commented 3 years ago

I am trying to use the Gazebo simulation of A1 robot in Gazebo(the one in Champ package) but the robot fall over(I have added the desired properties for Gazebo in URDF).

The gait planning of the robot is working fine here the robot can be seen being teleoperated in RVIZ. But when the same robot is taken to Gazebo it is not able to teleoperate and falls after some time. To me there can be few reasons setting proper physics parameters, setting PID values for ros_control or may be something else which at the moment I don't know.

Anyone who can guide/point out some problems with the simulation.

grassjelly commented 3 years ago

Hey @tahir1069 this might be related to this issue https://github.com/chvmp/champ/issues/14

grassjelly commented 3 years ago

Hey @tahir1069 . Did you manage to resolve this?

tahir1069 commented 3 years ago

Hi @grassjelly I was able to solve the problem.

grassjelly commented 3 years ago

hey @tahir1069 could you please share how you resolved it for other users' reference? Thanks!

tahir1069 commented 3 years ago

Well I solved the issue by changing the values for the PID and exactly following the steps in above mentioned issue. The default one generated by champ also work. One issue for me was the laptop power, by changing to higher performance machine I was able to solve the problem. Because on a lower performance machine feedback and physics calculation are hardly synchronized so a lack in this was also the problem. The package for A1 can be found here.

Supergithubber commented 2 years ago

Well I solved the issue by changing the values for the PID and exactly following the steps in above mentioned issue. The default one generated by champ also work. One issue for me was the laptop power, by changing to higher performance machine I was able to solve the problem. Because on a lower performance machine feedback and physics calculation are hardly synchronized so a lack in this was also the problem. The package for A1 can be found here.

Hi！ I have also encountered the same problem, but the url you sent is not accessible, can you upload it to GIhub again? Thanks！