Improving autonomous vehicle in‐traffic safety using learning‐based action governor. Issue 2 (28th March 2022)

• Record Type: Journal Article
• Authors: Han, Kyoungseok; Li, Nan; Tseng, Eric; Filev, Dimitar; Kolmanovsky, Ilya; Girard, Anouck
• Is Part Of: Advanced control for applications. Volume 4:Issue 2(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-03-28
• Subjects: action governor -- autonomous vehicle -- learning‐based control -- reinforcement learning; Automatic control -- Periodicals; Automatic control; Periodicals; Electronic journals; 629.8
• DOI: 10.1002/adc2.101
• Languages: English
• ISSNs: 2578-0727

Abstract

Abstract: The Action Governor (AG) is a supervisory scheme augmenting a nominal control system in order to enhance the system's safety and performance. It acts as an action filter, monitoring the action commands generated by the nominal control policy and adjusting the ones that might lead to undesirable system behavior. In this article, we present an approach based on learning to developing an AG for autonomous vehicle (AV) decision policies to improve their safety for operating in mixed‐autonomy traffic (i.e., traffic involving both AVs and human‐operated vehicles (HVs)). To achieve this, we demonstrate that it is possible to train the AG in a traffic simulator that is capable of representing in‐traffic interactions among AVs and HVs. We illustrate the effectiveness of this learning‐based AG approach to improving AV in‐traffic safety through simulation case studies. Abstract : The Figure shows a learning‐based approach to designing a safety supervisor for improving AV in‐traffic safety based on the Action Governor (AG) framework. The AG is an add‐on scheme augmenting a nominal control loop to enhance the system's safety and performance. The learning approach is based on identification and recording of unviable state and action pairs rather than only the ones that cause immediate safety violations.