FUTUR. Website for the scientific production of UPC researchers.

A deep reinforcement learning approach for path following on a quadrotor

Author: Rubi, B.; Morcego, B.; Perez, R.

Type of activity: Presentation of work at congresses
Name of edition: 2020 European Control Conference
Date of publication: 2020
Presentation's date: 2020-05-14
Book of congress proceedings: Proceedings of the 2020 European Control Conference (ECC): Saint Petersburg, Russia, May 12-15, 2020
First page: 1092
Last page: 1098
Project funding: Intelligent Supervision for the Predictive Maintenance of Industrial Systems
Safety and control of autonomous vehicles

Repository: http://hdl.handle.net/2117/328906
https://www.researchgate.net/publication/341384886_A_Deep_Reinforcement_Learning_Approach_for_Path_Following_on_a_Quadrotor
URL: https://ieeexplore.ieee.org/document/9143591
Abstract: This paper proposes the Deep Deterministic Policy Grandient (DDPG) reinforcement learning algorithm to solve the path following problem in a quadrotor vehicle. This agent is implemented using a separated control and guidance structure with an autopilot tracking the attitude and velocity commands. The DDPG agent is implemented in python and it is trained and tested in the RotorS-Gazebo environment, a realistic multirotor simulator integrated in ROS. Performance is compared with Adaptive NLGL, a g...

This paper proposes the Deep Deterministic Policy Grandient (DDPG) reinforcement learning algorithm to solve the path following problem in a quadrotor vehicle. This agent is implemented using a separated control and guidance structure with an autopilot tracking the attitude and velocity commands. The DDPG agent is implemented in python and it is trained and tested in the RotorS-Gazebo environment, a realistic multirotor simulator integrated in ROS. Performance is compared with Adaptive NLGL, a geometric algorithm that implements an equivalent control structure. Results show how the DDPG agent is able to outperform the Adaptive NLGL approach while reducing its complexity. © 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.
Citation: Rubi, B.; Morcego, B.; Perez, R. A deep reinforcement learning approach for path following on a quadrotor. A: European Control Conference. "Proceedings of the 2020 European Control Conference (ECC): Saint Petersburg, Russia, May 12-15, 2020". 2020, p. 1092-1098. ISBN 978-3-907144-02-2.
Keywords: Attitude control, Heuristic algorithms, Learning (artificial intelligence), Machine learning, Prediction algorithms, Training, Unmanned aerial vehicles

Group of research: CS2AC-UPC - Supervision, Safety and Automatic Control

Participants

Rubí Perelló, Bartomeu (author and speaker )
Morcego Seix, Bernardo (author and speaker )
Pérez Magrané, Ramon (author and speaker )

Attachments

ECC20_0016_FI.pdf (780896 bytes)