Tiny Robot Reinforcment Learning
A2R Lab
I’m Habeeba Mansour
- Senior
- CS and Chemistry major
- Do research at A2R Lab
Project Background
Bittle doing something cute!
- Realistic locomotion in robots has been a long-standing challenge
- Robots with a high number of degrees of freedom and multiple sensors have had success with locomotion
- Robots like Bittle with only 8 degrees of freedom and fewer sensors have not yet been able to achieve the same success
- We aim to train the Bittle robot to walk using reinforcement learning
Wins
NIVIDIA's Isaac Sim
- Found out everything NIVIDIA’s Isaac Sim doesn’t work on
- Scoped out different simulators
- Preparing algorithms to run once Isaac Sim is running
Plans for the rest of the semester
- Run RL algorithms on Isaac Sim
- Use transformers
Current Problems
- Still waiting on BCIT for the computer fix