Designing for the Extremes: Modeling Drivers’ Response Time to Take Back Control From Automation Using Bayesian Quantile Regression

OBJECTIVE Understanding the factors that affect drivers' response time in takeover from automation can help guide the design of vehicle systems to aid drivers. Higher quantiles of the response time distribution might indicate a higher risk of an unsuccessful takeover. Therefore, assessments of these systems should consider upper quantiles rather than focusing on the central tendency. BACKGROUND Drivers' responses to takeover requests can be assessed using the time it takes the driver to take over control. However, all the takeover timing studies that we could find focused on the mean response time. METHOD A study using an advanced driving simulator evaluated the effect of takeover request timing, event type at the onset of a takeover, and visual demand on drivers' response time. A mixed effects model was fit to the data using Bayesian quantile regression. RESULTS Takeover request timing, event type that precipitated the takeover, and the visual demand all affect driver response time. These factors affected the 85th percentile differently than the median. This was most evident in the revealed stopped vehicle event and conditions with a longer time budget and scenes with lower visual demand. CONCLUSION Because the factors affect the quantiles of the distribution differently, a focus on the mean response can misrepresent actual system performance. The 85th percentile is an important performance metric because it reveals factors that contribute to delayed responses and potentially dangerous outcomes, and it also indicates how well the system accommodates differences between drivers.

[1]  Nikolaus Umlauf,et al.  A primer on Bayesian distributional regression , 2018 .

[2]  Torrin M. Liddell,et al.  Bayesian data analysis for newcomers , 2018, Psychonomic bulletin & review.

[3]  Yulan Liang,et al.  A Looming Crisis , 2014 .

[4]  Riender Happee,et al.  Modeling take-over performance in level 3 conditionally automated vehicles. , 2017, Accident; analysis and prevention.

[5]  Alexandra Neukum,et al.  The effect of urgency of take-over requests during highly automated driving under distraction conditions , 2014 .

[6]  J C F de Winter,et al.  Comparing spatially static and dynamic vibrotactile take-over requests in the driver seat. , 2017, Accident; analysis and prevention.

[7]  Klaus Bengler,et al.  How Traffic Situations and Non-Driving Related Tasks Affect the Take-Over Quality in Highly Automated Driving , 2014 .

[8]  Heikki Summala,et al.  Driver/Vehicle Steering Response Latencies , 1981 .

[9]  Paul-Christian Bürkner,et al.  Advanced Bayesian Multilevel Modeling with the R Package brms , 2017, R J..

[10]  Stephen R. Porter,et al.  Quantile Regression: Analyzing Changes in Distributions Instead of Means , 2015 .

[11]  Kathrin Zeeb,et al.  Is take-over time all that matters? The impact of visual-cognitive load on driver take-over quality after conditionally automated driving. , 2016, Accident; analysis and prevention.

[12]  R. Baayen,et al.  Mixed-effects modeling with crossed random effects for subjects and items , 2008 .

[13]  J. Townsend,et al.  An accuracy-response time capacity assessment function that measures performance against standard parallel predictions. , 2012, Psychological review.

[14]  John D. Lee,et al.  Glances That Matter: Applying Quantile Regression to Assess Driver Distraction from Off-Road Glances , 2018, Proceedings of the Human Factors and Ergonomics Society Annual Meeting.

[15]  R. Whelan.  Effective Analysis of Reaction Time Data , 2008 .

[16]  Joost C. F. de Winter,et al.  Determinants of take-over time from automated driving: A meta-analysis of 129 studies , 2019, Transportation Research Part F: Traffic Psychology and Behaviour.

[17]  Partha Deb,et al.  Twopm: Two-Part Models , 2015 .

[18]  Josef F. Krems,et al.  Keep Your Scanners Peeled , 2016, Hum. Factors.

[19]  Paul-Christian Bürkner,et al.  brms: An R Package for Bayesian Multilevel Models Using Stan , 2017 .

[20]  Natasha Merat,et al.  Highly Automated Driving, Secondary Task Performance, and Driver State , 2012, Hum. Factors.

[21]  Wendy Ju,et al.  Timing of unstructured transitions of control in automated driving , 2015, 2015 IEEE Intelligent Vehicles Symposium (IV).

[22]  J. Townsend,et al.  Spatio-temporal properties of elementary perception: an investigation of parallel, serial, and coactive theories , 1995 .

[23]  Neville A. Stanton,et al.  Takeover Time in Highly Automated Vehicles: Noncritical Transitions to and From Manual Control , 2017, Hum. Factors.

[24]  Denis Fize,et al.  Speed of processing in the human visual system , 1996, Nature.

[25]  Natasha Merat,et al.  Transition to manual: driver behaviour when resuming control from a highly automated vehicle , 2014 .

[26]  Alan Feingold,et al.  Sex Differences in Variability in Intellectual Abilities: A New Look at an Old Controversy , 1992 .

[27]  Franz J. Neyer,et al.  A Gentle Introduction to Bayesian Analysis: Applications to Developmental Research , 2013, Child development.

[28]  Natasha Merat,et al.  Coming back into the loop: Drivers' perceptual-motor performance in critical events after automated driving. , 2017, Accident; analysis and prevention.

[29]  Zhenji Lu,et al.  How much time do drivers need to obtain situation awareness? A laboratory-based study of automated driving. , 2017, Applied ergonomics.

[30]  Christopher D. Wickens,et al.  In-Vehicle Glance Duration , 2007 .

[31]  Joshua E. Domeyer,et al.  Assessing Drivers’ Trust of Automated Vehicle Driving Styles With a Two-Part Mixed Model of Intervention Tendency and Magnitude , 2019, Hum. Factors.

[32]  J T Townsend,et al.  Truth and consequences of ordinal differences in statistical distributions: toward a theory of hierarchical inference. , 1990, Psychological bulletin.

[33]  Neville A. Stanton,et al.  The Development of a Method to Assess the Effects of Traffic Situation and Time Pressure on Driver Information Preferences , 2015, HCI.

[34]  Kathrin Zeeb,et al.  What determines the take-over time? An integrated model approach of driver take-over after automated driving. , 2015, Accident; analysis and prevention.

[35]  J. D. Lee,et al.  On the scatter in creep rupture times , 1990 .

[36]  John D. Lee,et al.  After Vehicle Automation Fails: Analysis of Driver Steering Behavior after a Sudden Deactivation of Control , 2018 .

[37]  Keming Yu,et al.  Bayesian quantile regression , 2001 .

[38]  Jiqiang Guo,et al.  Stan: A Probabilistic Programming Language. , 2017, Journal of statistical software.

[39]  Neville A. Stanton,et al.  Take-Over Time in Highly Automated Vehicles , 2018, Driver Reactions to Automated Vehicles.

[40]  Riender Happee,et al.  Take-over performance in evasive manoeuvres. , 2017, Accident; analysis and prevention.

[41]  Daniel V. McGehee,et al.  Human Performance Models and Rear-End Collision Avoidance Algorithms , 2001, Hum. Factors.