Main content

A Federated Reinforcement Learning Approach for Autonomous Vehicle Platooning

Show full item record

Title: A Federated Reinforcement Learning Approach for Autonomous Vehicle Platooning
Author: Boin, Christian
Department: School of Engineering
Program: Engineering
Advisor: Lei, Lei
Abstract: Since 2016 federated learning (FL) has been an evolving topic of discussion in the artificial intelligence (AI) research community. In particular, the successful applications of FL in supervised learning applications led to the development and study of federated reinforcement learning (FRL). At the time of writing this thesis, few works exist on the topic of FRL applied to autonomous vehicle (AV) platoons. Furthermore, most FRL works implement either weight or gradient aggregation. We explore FRL’s effectiveness as a means to improve AV platooning by designing and implementing an FRL framework atop of a custom AV platoon environment. We study the application of FRL in AV platooning for two scenarios: (1) Inter-platoon FRL (Inter-FRL) – FRL applied across platoons; (2) Intra-platoon FRL (Intra-FRL) – FRL applied within platoons. It is concluded that Intra-FRL using weight aggregation (Intra-FRLWA) provides optimal performance for controlling an AV platoon, even with variable platoon length.
Date: 2022-07
Rights: Attribution 4.0 International
Related Publications: Boin C, Lei L, Yang SX. AVDDPG – Federated reinforcement learning applied to autonomous platoon control. Intell Robot 2022;2(2):145-67.

Files in this item

Files Size Format View
Boin_Christian_202207_MASc.pdf 2.793Mb PDF View/Open

This item appears in the following Collection(s)

Show full item record

Attribution 4.0 International Except where otherwise noted, this item's license is described as Attribution 4.0 International