LPV control for autonomous vehicles using a machine learning-based tire pressure estimation
Autor: | Tamas Hegedus, Dániel Fényes, Damien Koenig, Balázs Németh, Olivier Sename, Péter Gáspár |
---|---|
Přispěvatelé: | Institute for Computer Science and Control [Budapest] (SZTAKI), Hungarian Academy of Sciences (MTA), GIPSA - Safe, Controlled and Monitored Systems (GIPSA-SAFE), GIPSA Pôle Automatique et Diagnostic (GIPSA-PAD), Grenoble Images Parole Signal Automatique (GIPSA-lab), Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Centre National de la Recherche Scientifique (CNRS)-Université Grenoble Alpes (UGA)-Institut polytechnique de Grenoble - Grenoble Institute of Technology (Grenoble INP ), Université Grenoble Alpes (UGA)-Grenoble Images Parole Signal Automatique (GIPSA-lab), Université Grenoble Alpes (UGA) |
Jazyk: | angličtina |
Rok vydání: | 2020 |
Předmět: |
0301 basic medicine
ComputingMilieux_THECOMPUTINGPROFESSION business.industry Computer science Control (management) Scheduling (production processes) Tire pressure computer.software_genre Machine learning Simulation software [SPI.AUTO]Engineering Sciences [physics]/Automatic 03 medical and health sciences Variable (computer science) 030104 developmental biology 0302 clinical medicine Control system Artificial intelligence business computer 030217 neurology & neurosurgery |
Zdroj: | 2020 28th Mediterranean Conference on Control and Automation (MED) MED 2020-28th Mediterranean Conference on Control and Automation MED 2020-28th Mediterranean Conference on Control and Automation, Sep 2020, Saint-Raphaël, France. ⟨10.1109/MED48518.2020.9183106⟩ MED |
Popis: | International audience; The paper presents a data-driven method for tire pressure estimation and an LPV-based control design for autonomous vehicles. The motivation of the research is that the pressures of the tires have high impacts on the lateral dynamics of the vehicle, because the loss of tire pressure may result in degradation in the lateral vehicle motion. First, a machine learning-based estimation algorithm, which uses only signals of on-board sensors, is proposed. Second, an LPV-based lateral control design is proposed, which uses the estimated tire pressure as a scheduling variable. The control is able to handle situations, in which the tire pressure decreases. The efficiency and the operation of the control system is illustrated through a comprehensive simulation example using the high-fidelity simulation software CarMaker. |
Databáze: | OpenAIRE |
Externí odkaz: |