Long-Term Planning with Deep Reinforcement Learning on Autonomous Drones

In this paper, we study a long-term planning scenario that is based on drone racing competitions held in real life. We conducted this experiment on a framework created for “Game of Drones: Drone Racing Competition” at NeurIPS 2019. The racing environment was created using Microsoft's AirSim Drone Racing Lab. We have trained a reinforcement learning agent, simulated quadrotor in our case, with the Policy Proximal Optimization (PPO) algorithm. After training process it was successfully able to compete against another simulated quadrotor that was running a classical path planning algorithm. Agent observations consist of data from IMU sensors, GPS coordinates of drone obtained through simulation and opponent drone GPS information. Using opponent drone GPS information during training helps dealing with complex state spaces which serves as expert guidance. This approach allows efficient and stable training process. Our work fits into Clought's level 10 UAV autonomy categorization. All experiments performed in this paper can be found and reproduced with code at our GitHub repository.