Németh, Balázs and Lelkó, Attila and Gáspár, Péter (2025) A framework for improving LPV control performance level with reinforcement learning. IFAC PAPERSONLINE, 59 (15). pp. 121-126. ISSN 2405-8971 10.1016/j.ifacol.2025.10.068

Text Nemeth_121_36461380_ny.pdf Download (636kB)

Abstract

This paper presents the approach of hierarchical framework that is developed for improving the performance level of Linear Parameter-Varying (LPV) control systems. The improvement is achieved through learning methods and the hierarchical framework is built-in the training process. An overview on the design method is introduced, focusing on the performance improvement of controller within the loop. The provided framework incorporates three elements, such as the learning-based agent, the LPV-based control, and the supervisor that guarantees the coordination between the previous two elements. The paper presents an application possibility of the methods, focusing on the design of autonomous vehicle control. © © 2025 The Authors.

Item Type:	Article
Uncontrolled Keywords:	design; Learning systems; reinforcement learning; reinforcement learning; Control system synthesis; Autonomous agents; Training process; Design method; Personnel training; Control performance; Learning methods; Autonomous Vehicles; Autonomous Vehicles; Autonomous Vehicles; Linear Parameter Varying Control; Reinforcement learnings; learning and control; learning and control; Performance:level;
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:	28 Jan 2026 20:59
Last Modified:	28 Jan 2026 20:59
URI:	https://eprints.sztaki.hu/id/eprint/11105

Update Item