Training for Advanced Driver Assistance Systems
Motor vehicle crashes account for over 35,000 fatalities and 2 million injuries per year. In addition to injuries and fatalities, crashes are a significant financial burden: over $23 billion dollars are spent on medical injuries from motor vehicle crashes each year. As the majority of crashes are attributed to human error, Advanced Driver Assistance Systems (ADAS) hold much promise in terms of reducing crashes and their associated healthcare burden. However, automation related ADAS at SAE levels 1-3 require human supervision: when the ADAS fails or behaves unexpectedly, the driver must respond. Unfortunately, the driver may not be ready for such a situation. Training that exposes drivers to specific response situations may provide them with the experience needed to make appropriate and timely responses to unexpected ADAS situations. The role and importance of training in understanding and using partial vehicle automation is an emerging topic that needs to be better understood, as suggested by other researchers (Beggiato & Krems, 2013; Bianchi et al., 2014; Koustanaï et al, 2012; Larsson, 2012; Marcos, 2018).
To date, several studies have examined different aspects of training and ADAS. For example, Koustanaï et al (2012) showed that familiarizing drivers to forward collision warning (FCW) systems using a driving simulator improved their performance. Similarly, Payre et al (2016) conducted a study to investigate the effect of elaborated training on driver performance. They found that the response time to emergency take back control situations was less for those who received training. However, some studies show no or negative effects of training. For example, Manser et al (2018) designed and tested two training approaches: video-based training and demonstration-based training delivered by a human instructor. Results showed that there was no significant improvement in drivers’ headway time to vehicle ahead for either training program. Collectively, these outcomes suggest that more work is needed in this area.
The objectives are to: (1) assess which elements of training (i.e., content) regarding ADAS support superior consumer knowledge and understanding of ADAS and lead to more appropriate system use, (2) determine what medium or format of training leads to better outcomes, and (3) identify which agencies should develop and standardize ADAS training.
Objectives 1 and 2 will likely involve human subjects experiments in real world or simulated environments. Objective 3 might involve expert stakeholder interviews or focus groups.
Current ADAS only function on limited roadway types, within finite geographic areas, within certain speed ranges, and under precise environmental conditions. Past research has shown that drivers either misperceive or oversimplify ADAS capabilities and they only remember ADAS limitations if they experience them. Training programs that provide drivers with an experience of the system limitations and allow them to practice dealing with such limitations can prove effective as countermeasures to unexpected or hazardous ADAS behavior. As such, this project will serve as a necessary step in the overall development of a comprehensive system that prepares drivers for interactions with vehicles equipped with ADAS.
Beggiato, M., & Krems, J. F. (2013). The evolution of mental model, trust and acceptance of adaptive cruise control in relation to initial information. Transportation Research Part F: Traffic Psychology and Behaviour, 18, 47–57.
Bianchi Piccinini, G. F., Rodrigues, C. M., Leitão, M., & Simões, A. (2014). Reaction to a critical situation during driving with Adaptive Cruise Control for users and non-users of the system. Safety Science, 72, 116–126. https://doi.org/10.1016/j.ssci.2014.09.008
Koustanaï, A., Cavallo, V., Delhomme, P., & Mas, A. (2012). Simulator training with a forward collision warning system: Effects on driver-system interactions and driver trust. Human Factors, 54(5), 709–721.
Larsson, A. F. L. (2012b). Driver usage and understanding of adaptive cruise control. Applied Ergonomics, 43(3), 501–506.
Marcos, I. S. (2018). Challenges in Partially Automated Driving: A Human Factors Perspective (Vol. 741). Linköping University Electronic Press.
Manser, M., Klauer, S. G., & Machiani, S. G. (2018). Driver Training for Automated Vehicle Technology. Retrieved from https://www.vtti.vt.edu/utc/safe-d/index.php/projects/driver-training-for-automated-vehicle-technology/
Payre, W., Cestac, J., & Delhomme, P. (2016). Fully Automated Driving: Impact of Trust and Practice on Manual Control Recovery. Human Factors, 58(2), 229–241. https://doi.org/10.1177/0018720815612319
RB Behavioral Traffic Safety Cooperative Research Program; USDOT Intelligent Transportation Systems Program; USDOT University Transportation Centers
Research is appropriate for MS or Ph.D. research
|Sponsoring Committee:||ACH30, Human Factors of Vehicles
|Research Period:||12 - 24 months|
|RNS Developer:||Shannon C. Roberts, University of Massachusetts Amherst|
|Index Terms:||Driver support systems, Level 1 driving automation, Level 2 driving automation, Level 3 driving automation, Driver training, |
Safety and Human Factors
Education and Training
Vehicles and Equipment