ai · Training
SpiRobs reach policy
A reinforcement-learning agent that steers a three-tendon arm to a target
Project page: SpiRobs
- Simulator
- MuJoCo, 3D
- Task
- Reach, tip to a target point
- Actuation
- 3 tendon actuators
- Method
- Reinforcement learning, training in progress
The environment and the policy are trained on the SpiRobs simulation. Curves, success rates and a video of the trained arm will go here once I have numbers I trust.