A Udacity project implementation. The task was to train a Quadcopter to fly using a reinforcement learning algorithm.
The project contains a reinforcment learning agent and training implementation and a jupyter file which reports the results of training.
The agent is implemented using the Deep Deterministic Policy Gradient algorithm - https://arxiv.org/pdf/1509.02971.pdf
The project isn't really intended for use, although the jupyter project file can be re-run. Rather the intent is to visualise the outcome of the project in the final jupyter file as a report.