Projects / Programmes source: ARIS

Neurophysiological and cognitive profiling of driving skills

Research activity

Code Science Field Subfield
2.08.00  Engineering sciences and technologies  Telecommunications   

Code Science Field
T180  Technological sciences  Telecommunication engineering 

Code Science Field
2.02  Engineering and Technology  Electrical engineering, Electronic engineering, Information engineering 
driving, distraction, behavior model, skills, machine learning, profiling
Evaluation (rules)
source: COBISS
Researchers (13)
no. Code Name and surname Research area Role Period No. of publicationsNo. of publications
1.  37830  Marina Horvat  Psychology  Researcher  2017 - 2020  82 
2.  29554  PhD Grega Jakus  Telecommunications  Researcher  2017 - 2020  105 
3.  16140  PhD Anton Kos  Telecommunications  Researcher  2017 - 2018  288 
4.  15365  PhD Andrej Košir  Telecommunications  Researcher  2018  325 
5.  21809  PhD Bojan Musil  Psychology  Researcher  2017 - 2020  353 
6.  52215  Nejc Plohl  Psychology  Researcher  2019 - 2020  93 
7.  23408  PhD Jaka Sodnik  Telecommunications  Head  2017 - 2020  298 
8.  30413  PhD Sara Stančin  Telecommunications  Researcher  2017 - 2020  49 
9.  37510  PhD Kristina Stojmenova Pečečnik  Telecommunications  Researcher  2018 - 2020  95 
10.  23347  PhD Gregor Strle  Computer science and informatics  Researcher  2019  66 
11.  33802  PhD Sara Tement  Psychology  Researcher  2017 - 2020  267 
12.  04148  PhD Sašo Tomažič  Telecommunications  Researcher  2017 - 2020  508 
13.  05934  PhD Anton Umek  Telecommunications  Researcher  2017 - 2018  194 
Organisations (2)
no. Code Research organisation City Registration number No. of publicationsNo. of publications
1.  1538  University of Ljubljana, Faculty of Electrical Engineering  Ljubljana  1626965  27,578 
2.  2565  University of Maribor Faculty of Arts  Maribor  5089638050  33,040 
Currently the most common basis for driver behavior modeling and assessment of driving skills is derived from self-reporting psychological tests and statistical analyses of past accidents and accident-causing behavior. The major problem with the use of accidents as a basis for evaluating individual drivers is their infrequency and the lack of reliability of their record.   The main goal of the proposed project is to develop a system for holistic evaluation of driving skills and to propose an improved driver behavior model based on physical performance and psychophysiological, biometrical and neuropsychological responses. The proposed system will be built around an advanced motion-based driving simulator providing immersive driving experience in a controlled and safe environment. The latter will enable development of a variety of driving scenarios with traffic including cars, trucks, cyclists, pedestrians, etc. These scenarios will include normal driving conditions with different varying traffic, road and weather conditions as well as a variety of unusual, stressful and even hazardous situations which cannot be duplicated in real life. Speeds, accelerations, steering control, lane deviations, safety distances, reaction times and other properties will be measured and logged. In addition to data collected within the simulation a set of external psychological, biometrical and neurophysiological instrumentation will also be used to assess driver’s psychophysical responses to stress and high cognitive workload.   A proprietary backend solution will be proposed to store all captured data as a unique dataset available for advanced machine learning algorithms and modeling. The latter will include various high-level feature engineering and extraction techniques with the goal of proposing a general driver behavior model which will explain how drivers react in different situations on the road. It will enable also identification of various driver types and profiles as well as common response strategies in varying psychophysical conditions. Individual’s reactions in different situations and impact of stress and cognitive workload on driving performance will also be identified and their overall assessment of driving skills will be provided.   The feasibility of the project will be insured by collaboration of an interdisciplinary team from the fields of electrical engineering and psychology as well as three co-financing companies. The involvement of two high tech companies from the fields of driving simulation technologies and applied neuroscience will provide advanced high tech instrumentation and domain expertise. The involvement of a large insurance company on the other hand will provide valuable domain knowledge on traffic safety and records of dangerous and critical situations in real life as well as promotional support for dissemination of project results.
Significance for science
Many researchers have observed different biometrical and physical responses of drivers by exposing them to a variety of road and traffic situations as well as distracting factors occurring inside and outside of the vehicle. The result of the proposed project will enable a novel holistic approach to the study of driver behavior and evaluation of driver’s skills by combining psychophysical, biometrical, neurophysiological and physical responses to a unified and highly organized dataset. All responses will be collected in a simulated driving environment where typical as well as critical and hazardous situations can be created in a safe, controlled and repeatable manner. Such unified data collection from drivers will enable machine learning algorithms to explore new relations and interactions among individual factors, potential causalities in a way that could not be studied before.   The second research field also addressed with this project is the development of AI-based algorithms for autonomous vehicles. AI-based algorithms will sooner or later be challenged with different critical situations where instant reactions and responses will be required. These responses will often be a subject of ethical decisions where injuries or at least material damage will sometime be unavoidable. The data acquired through this project will contain a huge collection of human reactions and responses in critical situations in traffic where only human experience and intuition can successfully avoid heavy accidents and prevent severe injuries or even deaths. These datasets will represent a valuable input for decision making processes in these algorithms.   Another relevant impact of the project on development of science in Slovenia will be a transfer of knowledge and skills among the researchers from two different disciplines (telecommunications and psychology) and two different universities. The proposed project is expected to bring several scientific results that will have a potential for the definition of new research directions on both fields. For instance, driving safety is a prominent topic in psychology. However, psychological studies to date have mostly relied on self-report of driving performance and retrospective self-reports of accidents. The use of the refined NERVteh motion-based driving simulator will help to validate previous studies on the link between personality traits and executive functions and bring new insights on the psychological determinants of driving performance.   Although the project is of an applicative nature and is focused primarily on the technical problem of collecting and analyzing a large set of data, the results of the project, which will not interfere with the achieved competitive advantages of the co-funding companies, are foreseen to be published in the top ranking scientific literature.
Significance for the country
The ability of predicting behaviors of human drivers and profiling their driving skills will be a crucial element of future risk assessment in the domains of insurance policy making, transportation and infrastructure planning and also design of algorithms for autonomous vehicles,.   An insurance company has already expressed their interest in such technology and joined the project as one of the co-financers. The assessment of driving skills and tools for driver training, would allow observation of driving performance of their customers, especially organizations with large fleets of vehicles and professional drivers. Special focus will be given to young drivers who are the by default classified as the most critical and risky population in everyday traffic and consequentially penalized with high insurance policy fees. Young drivers with proven above average driving skills and behavior could therefore be rewarded with discounts before they reach the required driving experience set by default. Such holistic driver assessment and training tool could be also used in driving schools, where driver instructors could create psychophysiological profiles of future drivers. The latter would include the driving skills, attitudes, knowledge and understanding of how they can manage driving risks. These results could identify specific training needs for every individual driver in order to improve the identified lack of skills.   In the field of infrastructure planning such driving behavior assessment tool can be used to test new planed road sections and parts of infrastructure before they are built. It will be simple to create specific infrastructure sections in a simulated environment and evaluate driver’s responses in terms of comfort, stress, cognitive level required to perceive and understand new road signs, traffic lights, etc.   NERVteh, as the other co-financer of the project collaborates with different manufactures of autonomous vehicles and provides them with advanced simulation technologies. All manufacturers of autonomous vehicles confirmed that the major problem related to the development of advanced AI algorithms for their vehicles is how to handle critical and unpredicted traffic situations and in the environment. They believe these algorithms should include human factor and human patterns for decision making in such difficult situations. The results of this project will provide a big and systematically organized data sets of human behavior and reactions in many typical critical and dangerous situation on the road which could be used for improving these AI algorithms. The results will therefore have an essential value also for Slovenian companies and research institutions in this domain.
Most important scientific results Interim report, final report
Most important socioeconomically and culturally relevant results Interim report, final report
Views history