© 2024 SURF
Despite elaborate training and examination procedures, a significant number of road crashes and fatalities involve novice drivers not only in the Netherlands but also worldwide. Feedback is a very essential component of learning to drive, therefore, providing effective feedback would enhance the understanding of critical situations and minimize the likelihood of hazardous situations. During traditional driving lessons, drivers receive subjective feedback from the instructor in the real world but may contain cognitive and economic biases. Though quantitative feedback can be obtained from the driving simulator, it cannot match real-world scenarios. Thus the main objective of this project was to investigate methods to bring the best best of both worlds, i.e. quantitative feedback in real-world conditions. To be able to provide appropriate quantitative feedback, it is first important to study/measure the driving proficiency quantitatively. As a part of this KIEM project, two special manoeuvres, the reverse bay parking and a three-point turn were considered for investigation. Firstly, the project team carefully studied current practices. The KPI’s based on which the drivers are trained and evaluated can be broadly classified into two categories “vehicle control” and “viewing behaviour”. The digital equivalence of the KPI’s related to these categories were obtained using in-vehicle CAN data to monitor the vehicle control aspects and a camera to monitor the viewing behaviour of the driver. Subsequently, data from several participants with different levels of driving experiences were collected and were simultaneously evaluated by the driving instructor. These evaluations were used to label the data. Machine learning algorithms were developed and trained to identify patterns and features in the data related to each of the selected KPI's to quantitatively evaluate the novice drivers. Based on the results achieved during this project, the project team concludes that quantitatively measuring the driving performance of the learner driver using modern data sources is feasible.
Receiving the first “Rijbewijs” is always an exciting moment for any teenager, but, this also comes with considerable risks. In the Netherlands, the fatality rate of young novice drivers is five times higher than that of drivers between the ages of 30 and 59 years. These risks are mainly because of age-related factors and lack of experience which manifests in inadequate higher-order skills required for hazard perception and successful interventions to react to risks on the road. Although risk assessment and driving attitude is included in the drivers’ training and examination process, the accident statistics show that it only has limited influence on the development factors such as attitudes, motivations, lifestyles, self-assessment and risk acceptance that play a significant role in post-licensing driving. This negatively impacts traffic safety. “How could novice drivers receive critical feedback on their driving behaviour and traffic safety? ” is, therefore, an important question.
Due to major advancements in domains such as ICT, sensors, big data, and Artificial Intelligence (AI), in-vehicle data is being extensively used for monitoring driver behaviour, driving style identification and driver modelling. However, use of such techniques in pre-license driver training and assessment has not been extensively explored. EIDETIC aims at developing a novel approach by fusing multiple data sources such as in-vehicle sensors/data (to trace the vehicle trajectory), eye-tracking glasses (to monitor viewing behaviour) and cameras (to monitor the surroundings) for providing quantifiable and understandable feedback to novice drivers. Furthermore, this new knowledge could also support driving instructors and examiners in ensuring safe drivers. This project will also generate necessary knowledge that would serve as a foundation for facilitating the transition to the training and assessment for drivers of automated vehicles.