how to support our won robot, except unitree robot

[Bailey-24]

[HansZ8]

For adapting to your robot:

• Sim: You can simply add your robot’s XML file and provide a corresponding MujocoEnv config.
• Real: Re-use the same DoFConfig, then create a new Environment that calls your robot’s SDK.
  The relevant docs and code are here:
• https://github.com/HansZ8/RoboJuDo/blob/0e87b4f759800d0433fcd4ec1012bd48c5870d57/docs/environment.md
• https://github.com/HansZ8/RoboJuDo/blob/0e87b4f759800d0433fcd4ec1012bd48c5870d57/robojudo/config/g1/env/g1_mujuco_env_cfg.py

For the policies we’ve already supported, you’ll need to tweak the training code. For RoboJuDo deploy, in theory, you only need to adjust the config, since we’ve decoupled robot params in policy code.
check:

• https://github.com/HansZ8/RoboJuDo/blob/0e87b4f759800d0433fcd4ec1012bd48c5870d57/docs/policy.md
• https://github.com/HansZ8/RoboJuDo/blob/0e87b4f759800d0433fcd4ec1012bd48c5870d57/robojudo/config/g1/policy/g1_asap_policy_cfg.py