Budai, Csanád and Széles, Tamás and Németh, Balázs and Gáspár, Péter (2025) End-to-end Reinforcement Learning for Autonomous Racing: Bridging the sim-to-real gap. In: 2025 American Control Conference (ACC). Proceedings of the American Control Conference . IEEE, Piscataway (NJ), pp. 200-205. ISBN 9798331569372 10.23919/ACC63710.2025.11107738

Text BudaiCs_200_36355074_z.pdf Restricted to Registered users only Request a copy

Abstract

Deep reinforcement learning is a promising technique that can help create autonomous agents. However, it is still an open problem how one can create a controller with robust operation for real-world automotive systems. The difficulty lies in either sample efficiency for real-world learning or developing a good enough simulator for training. This paper addresses the latter, proposing a method that provides a solution to the sim-to-real gap through domain randomization, learning with disturbances, and observation preprocessing. The method is validated on a small-scale F1TENTH-type test vehicle, that is trained to race autonomously in a fully end-to-end manner. It is demonstrated that the training process results in a policy that can drive the car safely even over the grip limit. © 2025 AACC.

Item Type:	Book Section
Uncontrolled Keywords:	AGENTS; Small scale; Autonomous agents; Real-world; Personnel training; Robust operation; Automotive systems; Deep learning; Deep reinforcement learning; End to end; randomisation; Reinforcement learnings; Racing automobiles; Type tests; Real-world learning; Test vehicle;
Subjects:	Q Science > QA Mathematics and Computer Science > QA75 Electronic computers. Computer science / számítástechnika, számítógéptudomány
Divisions:	Systems and Control Lab
SWORD Depositor:	MTMT Injector
Depositing User:	MTMT Injector
Date Deposited:	20 Jan 2026 11:51
Last Modified:	20 Jan 2026 11:51
URI:	https://eprints.sztaki.hu/id/eprint/11067

Update Item