A3C for drone autonomous driving using Airsim

Abstract

In this work, we apply artificial intelligence to guide a drone to a certain point autonomously. Unreal engine creates a virtual environment where the drone can fly, and the algorithm is trained simulating the drone dynamics thanks to Airsim plugin. The implemented algorithm is Asynchronous Actor-Critic Advantage (A3C), which trains a neural network with less computing resources than standard reinforcement learning algorithms that normally needs costly GPUs. To prove these advantages, several experiments are run using a different number of parallel simulations (threads). The drone should reach a point randomly generated each episode. The reward, the value and the advantage function are used to evaluate the performance. As expected, these experiments show that a higher number of threads helps the leaning process improve and become more stable. These learning results are of interest to optimize the computing resources in future applications.

A Comprehensive Review of Recent Research Trends on Unmanned Aerial Vehicles (UAVs)

“…Aerial Informatics and Robotics Simulation [340] is an open source UAV simulator developed by Microsoft. It provides a realistic simulation environment for testing and developing UAV control algorithms, including the advanced physics-based modeling of the UAV and its environment [341][342][343]. The software is designed to be flexible and customizable, allowing users to experiment with different control strategies and sensor configurations.…”