/CS238-Automated-Aircraft-Landing-Reinforcement-Learning

Using a RL agent and Dynamic Programming ideas to teach an AI agent to land an aircraft.

Primary LanguagePython

CS238-Automated-Aircraft-Landing-Reinforcement-Learning

Using a RL agent and Dynamic Programming ideas to teach an AI agent to land an aircraft. learning_algorithm3.py : We make a RL agent that tries to solve the 2D and 3D - MDP for the optimal policy. dynamic_programming.py : We implement a Monte-Carlo based DP algorithm which we parallelize and run on a 140 core cluster. Const.py : A file which holds all the constants that we are using simulator.py : The simulator that we made for the AI agent to interact with