Backing Up Control of a Self-Driving Truck-Trailer Vehicle with Deep Reinforcement Learning and Fuzzy Logic

Eduardo Bejar, Antonio Moran

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

10 Scopus citations

Abstract

This paper presents a control strategy for autonomous truck-trailer systems based on deep reinforcement learning (Deep RL). The deep deterministic policy gradient (DDPG) is used for training the controller using a reward function defined in terms of the desired final state of the system. A fuzzy-logic approach is employed to avoid the truck-trailer jackknife state. Simulation results show that the designed controller exhibits similar performance to state-of-the-art controllers such as the linear-fuzzy controller but with a much simpler design process.

Original languageEnglish
Title of host publication2018 IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages202-207
Number of pages6
ISBN (Electronic)9781538675687
DOIs
StatePublished - 2 Jul 2018
Event2018 IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2018 - Louisville, United States
Duration: 6 Dec 20188 Dec 2018

Publication series

Name2018 IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2018

Conference

Conference2018 IEEE International Symposium on Signal Processing and Information Technology, ISSPIT 2018
Country/TerritoryUnited States
CityLouisville
Period6/12/188/12/18

Keywords

  • control systems
  • deep learning
  • reinforcement learning
  • self-driving vehicles

Fingerprint

Dive into the research topics of 'Backing Up Control of a Self-Driving Truck-Trailer Vehicle with Deep Reinforcement Learning and Fuzzy Logic'. Together they form a unique fingerprint.

Cite this