Full driving automation refers to the sustained and unconditional performance of the entire dynamic driving task and dynamic driving task fallback by an Automated Driving System (ADS). This implies that the ADS is expected to operate the vehicle under all driver-manageable on-road conditions, without any expectation that a user will respond to a request to intervene. Such a broad operational design domain includes those areas of the road network that are inherently hazardous, which is why the ability to detect them must be an integral part of ADS functionality. The aim of this paper is to give an Automated Driving System the means to identify these hazardous areas so that it can adapt its behaviour accordingly. Intelligent agents are tasked with the evaluation of the Automated Driving System's performance during operation and record any violation of safety constraints during periodically recurring travels. If the ADS is confronted to several hazardous situations in a given area of the road network, specific rules are applied to reduce the severity and the likelihood of occurrence of similar situations during later trips. Experiments were conducted in the driving simulation software SCANeR studio to illustrate the proposed approach.
Adaptability of automated driving systems to the hazardous nature of road networks
2017-10-01
4384758 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Road traffic node road section driving adaptability evaluation method for automatic driving
Europäisches Patentamt | 2023
|Driving control method and system using road surface adaptability
Europäisches Patentamt | 2020
|DRIVING CONTROL METHOD AND SYSTEM USING ROAD SURFACE ADAPTABILITY
Europäisches Patentamt | 2019
|Implications of Off-road Automation for On-road Automated Driving Systems
SAE Technical Papers | 2023
|Strategies for Automated Updating of Road Networks for Driving Simulations
Deutsches Zentrum für Luft- und Raumfahrt (DLR) | 2017
|