Use this URL to cite or link to this record in EThOS:
Title: Testing automated driving systems to calibrate drivers' trust
Author: Khastgir, Siddartha
ISNI:       0000 0004 9357 9655
Awarding Body: University of Warwick
Current Institution: University of Warwick
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Access from Institution:
Automated Driving Systems (ADSs) offer many potential benefits like improved safety, reduced traffic congestion and lower emissions. However, such benefits can only be realised if drivers trust and make use of such systems. The two challenges explored in this thesis are: 1) How to increase trust in ADSs? 2) How to identify the test scenarios to establish the true capabilities and limitations of ADSs? Firstly, drivers’ trust needs to be calibrated to the “appropriate” level to prevent misuse (due to over trust) or disuse (due to under trust) of the system. In this research, a method to calibrate drivers’ trust to the appropriate level has been created. This method involves providing knowledge of the capabilities and limitations of the ADSs to the driver. However, there is a need to establish the capabilities and limitations of the ADSs which form the knowledge to be imparted to the driver. Therefore, the next research contribution lies in the development of a novel method to establish the knowledge of capabilities and limitations of ADSs (used to calibrate trust) in a reliable manner. This knowledge can be created by testing ADSs. However, in literature, an unanswered research question remains: How to identify test scenarios which highlight the limitations of ADSs? In order to identify such test scenarios, a novel hazard based testing approach to establish the capabilities and limitations of ADSs is presented by extending STPA (a hazard identification method) to create test scenarios. To ensure reliability of the hazard classification (and of the knowledge), the author created a novel objective approach for risk classification by creating a rule-set for risk ratings. The contribution of this research lies in developing a method to increase trust in ADSs by creating reliable knowledge using hazard based testing approach which identifies how an ADS can fail.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: TL Motor vehicles. Aeronautics. Astronautics