Motion Planning by Reinforcement Learning for an Unmanned Aerial Vehicle in Virtual Open Space with Static Obstacles

TL;DR

This paper demonstrates the use of reinforcement learning with proximal policy optimization to enable a virtual UAV to navigate around static obstacles in open space, achieving an 81% success rate in reaching goals.

Contribution

It introduces a reinforcement learning approach for UAV motion planning in virtual environments, reducing real-world testing costs and demonstrating effective navigation.

Findings

Mean reward and goal rate increased during training

UAV reached the goal with 81% success rate

Reinforcement learning effectively enabled obstacle avoidance

Abstract

In this study, we applied reinforcement learning based on the proximal policy optimization algorithm to perform motion planning for an unmanned aerial vehicle (UAV) in an open space with static obstacles. The application of reinforcement learning through a real UAV has several limitations such as time and cost; thus, we used the Gazebo simulator to train a virtual quadrotor UAV in a virtual environment. As the reinforcement learning progressed, the mean reward and goal rate of the model were increased. Furthermore, the test of the trained model shows that the UAV reaches the goal with an 81% goal rate using the simple reward function suggested in this work.