panda-gym

Why we included this project

panda-gym puts a Franka Emika Panda arm inside PyBullet and exposes it as Gymnasium environments for reach, push, slide, pick-and-place, stack, and flip tasks. The continuous action and observation spaces plug straight into standard RL libraries, so you can test an algorithm without building a simulator first. It is light enough to run on a laptop, which makes it a sensible choice for coursework, research prototypes, or quick benchmarking before moving to heavier simulators or real hardware. The documentation is solid, and the project is a practical bridge between textbook reinforcement learning and real robotic manipulation.