Project implementation as a part of Reinforcement learning subject at university of Freiburg.
The project is based on the Cart Pole problem in continuous cartpole.py. This classic control task has a four-dimensional continuous observation space – position of the cart, velocity of the cart, angle of the pole, rotation rate of the pole – as well as a one-dimensional continuous action space in [−1, 1] representing the force which is applied to the cart. The goal is to push and hold the pole in an upright position.