balancebot_rl

Implemented DQN on a self balancing simulated bot using pybullet and openai gym, baselines. The environment is made using ROS URDF and DQN is an off-policy reinforcement learning algorithm which can be used for discrete-action space.

How to run:

1. Download the zip
2. Create a conda environment and run "pip install -e ."
3. To run : python balancebot_task.py