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

Abstract

Objective: This paper aims to describe and test novel computational driver models, predicting drivers’ brake reaction times (BRTs) to different levels of lead vehicle braking, during driving with cruise control (CC) and during silent failures of adaptive cruise control (ACC). Background: Validated computational models predicting BRTs to silent failures of automation are lacking but are important for assessing the safety benefits of automated driving. Method: Two alternative models of driver response to silent ACC failures are proposed: a looming prediction model, assuming that drivers embody a generative model of ACC, and a lower gain model, assuming that drivers’ arousal decreases due to monitoring of the automated system. Predictions of BRTs issued by the models were tested using a driving simulator study. Results: The driving simulator study confirmed the predictions of the models: (a) BRTs were significantly shorter with an increase in kinematic criticality, both during driving with CC and during driving with ACC; (b) BRTs were significantly delayed when driving with ACC compared with driving with CC. However, the predicted BRTs were longer than the ones observed, entailing a fitting of the models to the data from the study. Conclusion: Both the looming prediction model and the lower gain model predict well the BRTs for the ACC driving condition. However, the looming prediction model has the advantage of being able to predict average BRTs using the exact same parameters as the model fitted to the CC driving data. Application: Knowledge resulting from this research can be helpful for assessing the safety benefits of automated driving.

Predicting takeover response to silent automated vehicle failures

“…The present findings demonstrate that this pattern holds for silent failures on bending roads, across a wide range of failure criticalities. The non-unity increase could have implications for the perceptual mechanisms underpinning how drivers decide when to intervene in silent failures [7]. The perceptual error at response (quantified by lower TLC T values) decreased with more gradual failures.…”

“…Such behaviour could be explained by accounts of drivers responding to the accumulated perceptual error, equating integration of a small error over a long time with the integration of a large error over a short time, resulting in responses at smaller absolute error in less urgent situations (cf. [7,9,51]). Though TLC T increased for less critical failures, TLC T values decreased due to slower responses when drivers were engaged with the auditory cognitive task.…”

“…models that describe how perceptual inputs are related to control) to simulate driver behaviour and determine the situations that will be the most problematic. Piccinini et al [7] have had some success at computationally capturing braking reaction times during silent adaptive cruise control failures. Drivers had longer reaction times than when manually driving, and also longer reaction times for less critical failures.…”

Variable-drift diffusion models of pedestrian road-crossing decisions

“…We and others have investigated the application of drift diffusion-type models in the road traffic context, with promising results initially for low-level locomotion decisions on applying braking or steering control [39,48,71], more recently also extending to multi-agent interaction situations [2,24,31,41,73].…”

A Review of Driving Simulation Technology and Applications

“…This feature allows the driver to specify its maximum speed and minimum distance to the preceding vehicle. Examples of this work can be seen in [104] where driving simulation is used to explore the use of an ACC system in traffic situations, and in [105] where system failure is explored in different traffic scenarios.…”