Increasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning

This paper presents a novel approach for improving the safety of vehicles equipped with Adaptive Cruise Control (ACC) by making use of Machine Learning (ML) and physical knowledge. More exactly, we train a Soft Actor-Critic (SAC) Reinforcement Learning (RL) algorithm that makes use of physical knowl...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Sorin Liviu Jurj, Dominik Grundt, Tino Werner, Philipp Borchers, Karina Rothemann, Eike Möhlmann
Formato:	article
Lenguaje:	EN
Publicado:	MDPI AG 2021
Materias:	adaptive cruise control informed machine learning physics-guided reinforcement learning safety autonomous vehicles Technology T
Acceso en línea:	https://doaj.org/article/242fdc0348dd4dd2b7c1655c33fc7008
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:242fdc0348dd4dd2b7c1655c33fc7008
record_format	dspace
spelling	oai:doaj.org-article:242fdc0348dd4dd2b7c1655c33fc70082021-11-25T17:26:50ZIncreasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning10.3390/en142275721996-1073https://doaj.org/article/242fdc0348dd4dd2b7c1655c33fc70082021-11-01T00:00:00Zhttps://www.mdpi.com/1996-1073/14/22/7572https://doaj.org/toc/1996-1073This paper presents a novel approach for improving the safety of vehicles equipped with Adaptive Cruise Control (ACC) by making use of Machine Learning (ML) and physical knowledge. More exactly, we train a Soft Actor-Critic (SAC) Reinforcement Learning (RL) algorithm that makes use of physical knowledge such as the jam-avoiding distance in order to automatically adjust the ideal longitudinal distance between the ego- and leading-vehicle, resulting in a safer solution. In our use case, the experimental results indicate that the physics-guided (PG) RL approach is better at avoiding collisions at any selected deceleration level and any fleet size when compared to a pure RL approach, proving that a physics-informed ML approach is more reliable when developing safe and efficient Artificial Intelligence (AI) components in autonomous vehicles (AVs).Sorin Liviu JurjDominik GrundtTino WernerPhilipp BorchersKarina RothemannEike MöhlmannMDPI AGarticleadaptive cruise controlinformed machine learningphysics-guided reinforcement learningsafetyautonomous vehiclesTechnologyTENEnergies, Vol 14, Iss 7572, p 7572 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	adaptive cruise control informed machine learning physics-guided reinforcement learning safety autonomous vehicles Technology T
spellingShingle	adaptive cruise control informed machine learning physics-guided reinforcement learning safety autonomous vehicles Technology T Sorin Liviu Jurj Dominik Grundt Tino Werner Philipp Borchers Karina Rothemann Eike Möhlmann Increasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning
description	This paper presents a novel approach for improving the safety of vehicles equipped with Adaptive Cruise Control (ACC) by making use of Machine Learning (ML) and physical knowledge. More exactly, we train a Soft Actor-Critic (SAC) Reinforcement Learning (RL) algorithm that makes use of physical knowledge such as the jam-avoiding distance in order to automatically adjust the ideal longitudinal distance between the ego- and leading-vehicle, resulting in a safer solution. In our use case, the experimental results indicate that the physics-guided (PG) RL approach is better at avoiding collisions at any selected deceleration level and any fleet size when compared to a pure RL approach, proving that a physics-informed ML approach is more reliable when developing safe and efficient Artificial Intelligence (AI) components in autonomous vehicles (AVs).
format	article
author	Sorin Liviu Jurj Dominik Grundt Tino Werner Philipp Borchers Karina Rothemann Eike Möhlmann
author_facet	Sorin Liviu Jurj Dominik Grundt Tino Werner Philipp Borchers Karina Rothemann Eike Möhlmann
author_sort	Sorin Liviu Jurj
title	Increasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning
title_short	Increasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning
title_full	Increasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning
title_fullStr	Increasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning
title_full_unstemmed	Increasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning
title_sort	increasing the safety of adaptive cruise control using physics-guided reinforcement learning
publisher	MDPI AG
publishDate	2021
url	https://doaj.org/article/242fdc0348dd4dd2b7c1655c33fc7008
work_keys_str_mv	AT sorinliviujurj increasingthesafetyofadaptivecruisecontrolusingphysicsguidedreinforcementlearning AT dominikgrundt increasingthesafetyofadaptivecruisecontrolusingphysicsguidedreinforcementlearning AT tinowerner increasingthesafetyofadaptivecruisecontrolusingphysicsguidedreinforcementlearning AT philippborchers increasingthesafetyofadaptivecruisecontrolusingphysicsguidedreinforcementlearning AT karinarothemann increasingthesafetyofadaptivecruisecontrolusingphysicsguidedreinforcementlearning AT eikemohlmann increasingthesafetyofadaptivecruisecontrolusingphysicsguidedreinforcementlearning
_version_	1718412337605509120

Increasing the Safety of Adaptive Cruise Control Using Physics-Guided Reinforcement Learning

Ejemplares similares