Abstract
Current and foreseeable automated vehicles are not able to respond appropriately in all circumstances and require human monitoring. An experimental examination of steering automation failure shows that response latency, variability and corrective manoeuvring systematically depend on failure severity and the cognitive load of the driver. The results are formalised into a probabilistic predictive model of response latencies that accounts for failure severity, cognitive load and variability within and between drivers. The model predicts high rates of unsafe outcomes in plausible automation failure scenarios. These findings underline that understanding variability in failure responses is crucial for understanding outcomes in automation failures.
Highlights
Automated vehicles (AVs) are developing at a rapid pace, but designing a system that can safely respond to all scenarios within existing road infrastructure remains a huge challenge
Silent failures of automation can be classified based upon how quickly the driver would leave the road after the failure in the case of driver inaction (TLCF)
In the Introduction, we argue that metrics that are linked to the unfolding scenario should provide better indicators of safe takeover than reaction time, so the measure of detection is timeto-lane-crossing at takeover (TLCT)
Summary
Automated vehicles (AVs) are developing at a rapid pace, but designing a system that can safely respond to all scenarios within existing road infrastructure remains a huge challenge. There will be cases where the AV’s ability to drive safely and to monitor its performance, is impaired These scenarios can arise because the system has malfunctioned, reached a limitation it is not aware of, or unintentionally misclassifies or fails to classify an object in (or feature of) the road environment [e.g. the 2016 Tesla crash where the AV failed to identify a truck; [2]. There will be a “silent failure”, and it will be up to the supervising driver to detect that the AV has failed and to respond safely to the conditions Throughout this manuscript situations where the AV fails without providing any explicit alert to the driver will be referred to as silent failures (as per [3, 4]). Human detection of these silent failures in automated lane keeping, the resultant steering responses when regaining control, and how distraction affects these behaviours, will be the focus of this manuscript
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
