RC WheelLeg

What it can do

Written by Robotheca from the project's documents.

• Moves on four wheel-legs with 16 DOF using reinforcement learning policies
• Runs 50 Hz policy inference and 200 Hz motor control loop on Jetson Orin Nano
• Supports ROS 2 Navigation2 stack for waypoint navigation and mapping
• Provides safety monitoring, emergency stop and zero-command hold logic
• Publishes joint states, IMU data and runtime diagnostics over ROS 2 topics
• Accepts velocity commands via /cmd_vel for teleoperation or navigation

Works with AI

Found by Robotheca in the project's files at commit 1bc9aa5, its robot.yaml and Robotheca's notes on its documents. Never guessed; Robotheca doesn't run the code.

• Ships trained model files: model_6800.onnx and 13 more

  In the project files: 05_software/train/rc_mjlab/model_6800.onnx

• Uses reinforcement learning: rsl_rl

  In the project files: 05_software/train/rc_mjlab/src/robot/rsl_rl

• Uses reinforcement learning: mjlab

  In the project files: 05_software/train/rc_mjlab

• Uses a machine-learning library: PyTorch

  In the project files: 05_software/train/rc_mjlab/mjlab/src/mjlab/utils/torch.py

• Uses a machine-learning library: ONNX

  In the project files: 05_software/real/sim2real_v2/tools/export_onnx.py

• “Moves on four wheel-legs with 16 DOF using reinforcement learning policies”

  Robotheca's notes: What it can do

• “Download ONNXRuntime C++ SDK for Jetson Orin and extract to /opt/onnxruntime”

  Robotheca's notes: How to run it

• “Set ONNXRUNTIME_DIR=/opt/onnxruntime”

  Robotheca's notes: How to run it

Wiring and electronics

Notes written by Robotheca from the project's documents.