Developing Safe Strategies for Automated Vehicle Failures through Text Mining and Human Factors Methods
Associate Professor of Industrial and Manufacturing Systems Engineering, College of Engineering and Computer Science, the University of Michigan-Dearborn, Associate Research Scientist, Human Factors, University of Michigan Transportation Research Institute and Intermittent Lecturer in Industrial and Operations Engineering, College of Engineering
Associate Research Scientist, Engineering Systems, University of Michigan Transportation Research Institute
The recent rapid development of advanced driver assistance systems (ADAS) and automated driving systems (ADS) provides opportunities to improve road safety, but they can also create new safety challenges as those technologies all have limitations and are fallible. By examining owner manuals of 12 partially automated vehicles available on the market, Bao et al., (2019) have identified that the same ADAS function can have operation requirements that differ widely across different vehicle models, such as road type, speed limit, and other environmental factors like weather conditions. Without understanding the limitations of those ADAS and ADS functions, drivers will likely misuse the technology and cause failures and malfunctions of such systems.
The primary objectives of this project are to: 1) identify typical and important failure types and taxonomies for automated vehicle systems currently on the road; 2) examine and quantify the impact of relevant risk factors that are related to driver/operator responses (from both subject and surrounding vehicles) during a vehicle system failure; and 3) propose corresponding human factors coping strategies and design solutions in mitigating hazards of such vehicle failures and supporting safe and efficient responses for drivers from both subjects and surrounding vehicles. We propose a hybrid approach to address the research questions both qualitatively and quantitatively.