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

Modeling road user response timing in naturalistic traffic conflicts: A surprise-based framework

There is currently no established method for evaluating human response timing across a range of naturalistic traffic conflict types. Traditional notions derived from controlled experiments, such as perception-response time, fail to account for the situation-dependency of human responses and offer no clear way to define the stimulus in many common traffic conflict scenarios. As a result, they are not well suited for application in naturalistic settings. We present a novel framework for measuring and modeling response times in naturalistic traffic conflicts applicable to automated driving systems as well as other traffic safety domains. The framework suggests that response timing must be understood relative to the subject's current (prior) belief and is always embedded in, and dependent on, the dynamically evolving situation. The response process is modeled as a belief update process driven by perceived violations to this prior belief, that is, by surprising stimuli. The framework resolves two key limitations with traditional notions of response time when applied in naturalistic scenarios: (1) The strong situation dependence of response timing and (2) how to unambiguously define the stimulus. Resolving these issues is a challenge that must be addressed by any response timing model intended to be applied in naturalistic traffic conflicts. We show how the framework can be implemented by means of a relatively simple heuristic model fit to naturalistic human response data from real crashes and near crashes from the SHRP2 dataset and discuss how it is, in principle, generalizable to any traffic conflict scenario. We also discuss how the response timing framework can be implemented computationally based on evidence accumulation enhanced by machine learning-based generative models and the information-theoretic concept of surprise.

Is my AV crashing? An online photo-based experiment assessing whether shared intended pathway can help AV drivers anticipate silent failures

The shared responsibility between conditional AVs drivers demands shared understanding. Thus, a shared intended pathway (SIP)-a graphical display of the AV's planned manoeuvres in a head-up display to help drivers anticipate silent failures is proposed. An online, randomised photo experiment was conducted with 394 drivers in Australia. The photos presented traffic scenarios where the SIP forecast either safe or unsafe manoeuvres (silent failures). Participants were required to respond by selecting whether driver intervention was necessary or not. Additionally, the effects of presented object recognition bounding boxes which indicated whether a road user was recognised or not were also tested in the experiment. The SIP led to correct intervention choices 87% of the time, and to calibrating self-reported trust, perceived ease of use and usefulness. The bounding boxes found no significant effects. Results suggest SIPs can assist in monitoring conditional automation. Future research in simulator studies is recommended. Practitioner summary: Conditional AV drivers are expected to take-over control during failures. However, drivers are not informed about the AV's planned manoeuvres. A visual display that presents the shared intended pathway is proposed to help drivers mitigate silent failures. This online photo experiment found the display helped anticipate failures with 87% accuracy.

Applying the Accumulator model to predict driver's reaction time based on looming in approaching and braking conditions

INTRODUCTION

The prediction of when the driver will react to a change in the lead vehicle motion is critical for assessing rear-end crash risk using car-following models. Past studies have assumed constant reaction time and driver's continuous reaction. However, these assumptions are not valid as the driver's reaction time can vary in different car-following situations and the driver does not continuously react to the lead vehicle motion. Thus, this study predicted the driver's reaction time using the Wiedemann car-following model and the Accumulator model. The Accumulator model assumes the driver's start of reaction based on the accumulation of looming and thereby reflects the driver's intermittent reaction.

METHOD

Fifty drivers' behavior was observed using a driving simulator in two scenarios: (1) approach and follow a moving lead vehicle and (2) approach a stopped lead vehicle. The Accumulator model predicted the reaction times based on different looming variables (angular velocity and tau-inverse), lead vehicle type (car and truck), and lead vehicle brake lights (on or off).

RESULTS

The Accumulator model showed lower prediction errors of the reaction time than the Wiedemann model, which assumes reaction based on the fixed looming threshold. The Accumulator model predicted the reaction times more accurately when it was calibrated with the angular velocity due to width and height of lead vehicles. Moreover, the Accumulator model with tau-inverse produced the smallest prediction error of reaction times among different Accumulator models and the Wiedemann model when lead vehicle brake lights were on.

CONCLUSIONS

This study demonstrates that the Accumulator model is a promising method of predicting the driver's reaction time in car-following situations, which affects rear-end crash risk.

PRACTICAL APPLICATIONS

The Accumulator model can be incorporated into a car-following model for the prediction of reaction times and can estimate the rear-end collision risk of vehicles more accurately.

A Bayesian Regression Analysis of the Effects of Alert Presence and Scenario Criticality on Automated Vehicle Takeover Performance

OBJECTIVE

This study investigates the impact of silent and alerted failures on driver performance across two levels of scenario criticality during automated vehicle transitions of control.

BACKGROUND

Recent analyses of automated vehicle crashes show that many crashes occur after a transition of control or a silent automation failure. A substantial amount of research has been dedicated to investigating the impact of various factors on drivers' responses, but silent failures and their interactions with scenario criticality are understudied.

METHOD

A driving simulator study was conducted comparing scenario criticality, alert presence, and two driving scenarios. Bayesian regression models and Fisher's exact tests were used to investigate the impact of alert and scenario criticality on takeover performance.

RESULTS

The results show that silent failures increase takeover times and the intensity of posttakeover maximum accelerations and decrease the posttakeover minimum time-to-collision. While the predicted average impact of silent failures on takeover time was practically low, the effects on minimum time-to-collision and maximum accelerations were safety-significant. The analysis of posttakeover control interaction effects shows that the effect of alert presence differs by the scenario criticality.

CONCLUSION

Although the impact of the absence of an alert on takeover performance was less than that of scenario criticality, silent failures seem to play a substantial role-by leading to an unsafe maneuver-in critical automated vehicle takeovers.

APPLICATION

Understanding the implications of silent failure on driver's takeover performance can benefit the assessment of automated vehicles' safety and provide guidance for fail-safe system designs.

Computational modeling of driver pre-crash brake response, with and without off-road glances: Parameterization using real-world crashes and near-crashes

When faced with an imminent collision threat, human vehicle drivers respond with braking in a manner which is stereotypical, yet modulated in complex ways by many factors, including the specific traffic situation and past driver eye movements. A computational model capturing these phenomena would have high applied value, for example in virtual vehicle safety testing methods, but existing models are either simplistic or not sufficiently validated. This paper extends an existing quantitative driver model for initiation and modulation of pre-crash brake response, to handle off-road glance behavior. The resulting models are fitted to time-series data from real-world naturalistic rear-end crashes and near-crashes. A stringent parameterization and model selection procedure is presented, based on particle swarm optimization and maximum likelihood estimation. A major contribution of this paper is the resulting first-ever fit of a computational model of human braking to real near-crash and crash behavior data. The model selection results also permit novel conclusions regarding behavior and accident causation: Firstly, the results indicate that drivers have partial visual looming perception during off-road glances; that is, evidence for braking is collected, albeit at a slower pace, while the driver is looking away from the forward roadway. Secondly, the results suggest that an important causation factor in crashes without off-road glances may be a reduced responsiveness to visual looming, possibly associated with cognitive driver state (e.g., drowsiness or erroneous driver expectations). It is also demonstrated that a model parameterized on less-critical data, such as near-crashes, may also accurately reproduce driver behavior in highly critical situations, such as crashes.

Making a few talk for the many - Modeling driver behavior using synthetic populations generated from experimental data

Understanding driver behavior is the basis for the development of many advanced driver assistance systems, and experimental studies are indispensable tools for constructing appropriate driver models. However, the high cost associated with testing is a serious obstacle in collecting large amounts of experimental data. This paper presents a methodology that can improve the reliability of results from experimental studies with a limited number of participants by creating a virtual population. Specifically, a methodology based on Bayesian inference has been developed, that generates synthetic cases that adhere to various real-world constraints and represent possible variations of the observed experimental data. The application of the framework is illustrated using data collected during a test-track experiment where truck drivers performed a right turn maneuver, with and without a cyclist crossing the intersection. The results show that, based on the speed profiles of the dataset and physical constraints, the methodology can produce synthetic speed profiles during braking that mimic the original curves but extend to other realistic braking patterns that were not directly observed. The models obtained from the proposed methodology have applications for the design of active safety systems and automated driving, demonstrating thereby that the developed framework has great promise for the automotive industry.

Predicting takeover response to silent automated vehicle failures

Current and foreseeable automated vehicles are not able to respond appropriately in all circumstances and require human monitoring. An experimental examination of steering automation failure shows that response latency, variability and corrective manoeuvring systematically depend on failure severity and the cognitive load of the driver. The results are formalised into a probabilistic predictive model of response latencies that accounts for failure severity, cognitive load and variability within and between drivers. The model predicts high rates of unsafe outcomes in plausible automation failure scenarios. These findings underline that understanding variability in failure responses is crucial for understanding outcomes in automation failures.