Bipedal Walker - Reinforcement Learning

Image of Bipedal Walker - Reinforcement Learning

The goal of this project was to use reinforcement learning to train a physics-based agent (the bipedal walker) to maneuver over terrain and obstacles (the OpenAI gym environments).

We didn't reach our goal (reaching over 300 units of distance for 100 consecutive runs) but I got to make some fun GIFs out of it ^