About the Project
Booster Robotics originally provided reinforcement learning code using Isaac Gym to make the robots walk. Unfortunately, this framework is now deprecated.
I took on the task of implementing the Booster T1 into Isaac Lab in order to develop our own custom RL walking policies. This migration allowed for more modern simulation capabilities and better long-term support.
Currently, this is a work in progress. I am now looking at using MuJoCo Lab to further refine the policies and make the locomotion more robust in real-world environments.
Tech Stack Used
Key Features
- Modern Framework: Migrated from deprecated Isaac Gym to Isaac Lab.
- Custom Locomotion: Developed unique walking gaits tailored to the T1.
- Sim2Real Deployment: Validated physics simulation on real hardware.