Steer or Brake?: Modeling Drivers’ Collision-Avoidance Behavior by Using Perceptual Cues

Abstract

Driver models have been developed to capture collision-avoidance behaviors, yet there is a lack of understanding of what perceptual processes influence drivers’ choices to brake or steer. A statistical model of these decisions was developed with cluster analysis and multinomial logistic regression with data from a simulator study of drivers’ responses to rear-end collisions. Drivers’ choices of responses were clustered on the basis of the maximum values of the magnitude of braking and steering forces, starting from the time at which the driver looked back to the road, just before initiating the avoidance maneuver, to the end of the maneuver. The clusters identified three types of responses: medium to high braking with medium to high steering, medium to high braking with mild steering, and mild braking with medium to high steering. The perceptual variables such as the optical angle, the expansion rate of the optical angle, and their ratio were used to predict the drivers’ choice of response. The results show that, of the perceptual variables, the combination of optical angle and tau performs as well as or better than others in predicting the choice of response. The mode and timing of an alert from a collision-warning system did not influence the drivers’ choices. These results can inform driver behavior models to guide design and assess benefits of advanced driver assistance systems.

How Do Drivers Respond to Silent Automation Failures? Driving Simulator Study and Comparison of Computational Driver Braking Models

“…• The BRTs significantly decrease with higher levels of kinematic criticality, both during driving with CC and ACC. This outcome is in line with previous research (Markkula, 2014;Markkula et al, 2016;Engström et al, 2017;Venkatraman et al, 2016) but shows for the first time this phenomenon in silent failures of automation. • The BRTs are significantly longer during driving with ACC compared to driving with CC.…”

Decision Tree-Based Maneuver Prediction for Driver Rear-End Risk-Avoidance Behaviors in Cut-In Scenarios

“…Lastsecond braking and steering judgments in a car-following condition have been studied relative to the alert timing for forward collision warnings [17]. Focusing on the carfollowing scenario, a statistical model was developed to understand the perceptual processes involved in a driver's decision to brake or steer [18]. It was shown that, in imminent rear-end collision situations, most drivers tend only to brake and often do not attempt to avoid obstacles by steering.…”

“…A random-forest machine-learning technique was used to develop red-light-running violation-prediction models using a single objective indicator (time to intersection) [20]. Venkatraman et al applied cluster analysis and multinomial logistic regression to predict driver decisions in the carfollowing situation [18]. Generative approaches for maneuver prediction include various filtering methods, such as Kalman and particle filters, hidden Markov models, and dynamic Bayesian networks [21,22].…”

Modeling Drivers’ Strategy When Overtaking Cyclists in the Presence of Oncoming Traffic

“…In an overtaking scenario, being able to predict whether a driver would avoid collisions by braking (to initiate an accelerative maneuver) or by steering (to initiate a flying maneuver) can help active safety systems to tune intervention times [37], [38]. In fact, knowing a driver's overtaking strategy can also be valuable information for systems that aim at preventing a head-on collision with the oncoming traffic in the passing phase, as a result of a poorly timed flying maneuver [30], [32], and rear-end collisions with cyclists as a result of a poorly timed accelerative maneuver.…”