search

real-stanford / diffusion_policy

[RSS 2023] Diffusion Policy Visuomotor Policy Learning via Action Diffusion
https://diffusion-policy.cs.columbia.edu/
MIT License
1.09k stars 203 forks source link

wrong rotation of the predicted actions #65

Open yl-wang996 opened 2 months ago

yl-wang996 commented 2 months ago

As in the following figure, I try to visualize the predicted actions in the rviz. In this case, each prediction will generate 16 waypotins and the first 8 will be used. But I found the rotation part of the action is incorrect, there seems exist a fixed offset in the rotation. And the action is directly from here.

So my question is: is this rotation correct as in the following image, if not what could be the problem?

image

yl-wang996 commented 2 months ago

When I visualize the trajectories of the TCP, the rotation seems correct. image

yl-wang996 commented 2 months ago

And then I found the action in the demonstration dataset is also not vertical, is it correct? image

2292685327 commented 2 months ago

Hello, I am a beginner, how to use rviz for visualization?

yl-wang996 commented 1 month ago

Hello, I am a beginner, how to use rviz for visualization?

@2292685327 Basically, you just need to know how to use ROS, and for rviz you can refer to here

2292685327 commented 1 month ago

Hello, I am a beginner, how to use rviz for visualization?

@2292685327 Basically, you just need to know how to use ROS, and for rviz you can refer to here

Thank you for your prompt reply. I have another question. Are the robot files and required target files in the simulation environment already provided in the source code? Or do I need to recreate it myself?

yl-wang996 commented 1 month ago

@2292685327 sorry for the late reply, the robomimic rely on the robosuite, so the env is built by robosuite. An alternative is you can find the mujoco XML file in the attributes of the dataset file(***.h5py), but the path is wrong in this XML file.