Toy Cart-Pole

Repository for the Cart-Pole problem.

Implemented Classic Controllers:

• PID
• Pole-Placement in State-Space
• LQR
• MPC

To Do:

• Implement RL agent (discrete and continuous actions)

To run:

1. Create a venv
2. Clone the OpenAi Gym repo into project directory (Toy-CartPole/)
3. Add the /gym/gym/envs/classic_control/cartpole_cont.py file to the corresponding path
4. Add to /gym/gym/envs/__init__.py

register(
    id="CartPole-cont",
    entry_point="gym.envs.classic_control.cartpole_cont:CartPoleEnv",
    max_episode_steps=500,
    reward_threshold=475.0,
)

5. Do $pip install -e . inside /gym/