Although the capabilities of automated driving systems (ADS) are growing, the human operator remains in charge of driving the vehicle outside the operational design domain and after requests to intervene. The vehicle sensors required for ADS and new regulation for driver availability monitoring systems stimulate the development of systems that can monitor human driving ability. Such systems could allow for ADS to track possible driving skill degradation, monitor driver impairment, or adapt to the driver’s needs. However, knowledge is required about how to assess human driving ability effectively. This work proposes a novel method to capture driving skill and style based on curve driving, straight-road driving, and driving through road narrowings. In addition to conventional measures, the study introduces new measures, including the relationship between steering angle and eye movements. Employing a high-fidelity simulator, we compared the vehicle-control-related measures (i.e., driving skill measures), trajectory-planning and speed-related measures (i.e., driving style measures), and eye movements of sixteen inexperienced and thirteen experienced drivers. It was examined which of these measures are valid, i.e., allow a statistical discrimination between experienced and inexperienced drivers. The study was complemented with two expert drivers serving as a benchmark. The results showed that the experienced drivers adopted a more abrupt braking style when approaching curves and a higher speed through the different track sections than the inexperienced drivers. No statistically significant differences were observed between the skill measures of the experienced and inexperienced drivers. An analysis of lead times obtained through a cross-correlation between horizontal gaze angle and steering angle showed that eye movements generally preceded steering movements. However, differences in lead times between experienced and inexperienced drivers were not statistically significant, which may have been caused by eye-tracking measurement inaccuracies, the layout of the driving task and strong intra-individual variability in looking behavior. The eye-tracking results of the road-narrowings showed that the experienced and inexperienced drivers reduced their vertical gaze dispersion, while only the experienced drivers statistically significantly reduced their horizontal gaze dispersion compared to the straights. In other words, the experienced drivers showed increased horizontal gaze tunneling from the straights to the narrowings, while the inexperienced drivers only showed vertical gaze tunneling. The results of the expert drivers showed consistent higher speed, lower control activity, and more of a racing line through the curves than the experienced and inexperienced drivers. Furthermore, the results showed that the expert drivers adopted a more variable horizontal gaze strategy between different curves. Overall, these results indicate that driving style and eye-movement measures, but not driving skill measures, allow differentiating between experienced and inexperienced drivers using a driving simulator. These findings may be explained by the fact that driving is a self-paced task, i.e., more competent drivers increase their own task demands by driving faster. Future research could examine how strongly scores the present driving and eye-movement measures correlate with drivers’ take-over quality in automated driving scenarios.

Measuring cognitive load is essential for understanding driver performance. Under- and overload can result in dangerous situations on the road. Cognitive load can be estimated by monitoring the diameter and movements of the pupils, but during measurements external influences such as changes in light intensity affect pupil diameters. In this paper, we present a novel method for quantifying light intensity with a head-mounted eye-tracker by weighting pixel values around the gaze direction. We demonstrate its effectiveness in cognitive load classification systems that use pupil metrics only. 54 participants in two separate studies have carried out n-back tasks during a simple driving task in a driving simulator. The data is classified by cognitive task (baseline, 1-back, 2-back) with the Random Forest algorithm. The resulting systems are 92.5% accurate with and 85.9% accurate without gaze features available, but are unable to generalise to participants unseen in the training phase of the algorithm.

Due to recent advancements in automated driving systems, drivers are able to withdraw themselves from the control loop of the vehicle in certain driving situations. However, system limits occur relatively frequently, which results in the driver having to take back control. It is argued most currently used take-over quality quantification methods, such as identifying the minimum time-to-collision or maximum lateral acceleration, neglect inherent criticality of the driving scenario. This can lead to poor overall quality assessment, especially in challenging take-over scenarios. This thesis proposes a novel method of quantifying how well the driver performed in the task of executing an emergency automation-to-manual control takeover, by means of a scenario-specific quality reference in the form of an optimized trajectory. Firstly, a human-in-the-loop driving simulator study was carried out using a high fidelity 6-DOF driving simulator (n=25). The simulated environment consisted of a two-lane highway setting. During the drive, six different automation-to-manual takeover scenarios occurred. Optimized trajectories were also generated for these scenarios, using a State-of-the-Art trajectory planner developed at Volkswagen Group Research Facility, Wolfsburg. The independent variables were the time budget (5s, 7s, and 20s), and the traffic density (5 veh/km, and 10 veh/km). A variety of dependent variables were determined for both the human driver trajectories and the optimized trajectories: minimum time-to-collision, minimum distance-to-lane-boundary, minimum distance-to-overtaking-vehicle, maximum forced braking of the overtaking vehicle, maximum and standard deviations of accelerations (lateral and longitudinal), maximum and standard deviations of jerk (lateral and longitudinal), and lateral quickness. Based on the findings from this human-in-the-loop driving simulator study and the generated optimized trajectories, a quantification framework was developed. The framework subdivides take-over quality assessment into two separate parameters. One with regard to safety, and one with regard to comfort. The safety parameter quantifies criticality of the human driver take-over based on the sum of normalized weighted quantification scores. These scores are computed for the metrics minimum time-to-collision, minimum distance-to-lane boundary, minimum distance-to-overtaking vehicle, and maximum forced braking of the overtaking vehicle, and are based on the difference to the metric values corresponding to the optimized
trajectory. The comfort parameter quantifies comfort in a similar manner based on the metrics lateral quickness, maximum lateral and longitudinal jerk, and standard deviation of lateral and longitudinal jerk. The final quantification scores are comprised of a value between 0 and 1 for both the safety and comfort parameters. A perfect score of 0 would indicate all metrics were identical, or better than the optimized reference trajectory. A value of 1 indicates a crash for the safety parameter, and highly uncomfortable driving for the comfort parameter. The resulting quantification framework is scenario-specific as the quantification is performed relative to an optimized reference trajectory.