More than 1,350,000 people die in traffic every year. Automated vehicles (AVs) are expected to improve road safety. However, a challenge is that future traffic will be mixed, with AVs, vulnerable road users (VRUs), such as pedestrians and cyclists and motorcyclists, and manually-driven cars sharing the roads and pursuing conflicting interests. Also, AVs will have different forms, differing in capabilities, driving speed, operational design domain (ODD) and technological capabilities. Also, some vehicles may be privately owned, offering the driver to sometimes drive in AV-mode, or maybe another form of automated transport, such as robotaxis or automated shuttles, where the user is not the owner.
It is not clear how AVs can assess VRU's situation awareness (SA). According to the theory of Distributed Situation Awareness (DSA), safety arises through shared knowledge of the collective. However, no mechanism exists for assessing DSA in complex traffic, as DSA misses multidirectional information exchange between multiple agents, and the theory has not been tested in complex and connected traffic. Also, current empirical research on AV-VRU interactions focuses on either sending or receiving data, instead of multidirectional exchange. It is unknown how an AV should adapt to the knowledge provided by multiple vehicles and VRUs with different SA.
The aim of this PhD project will be to update DSA with the multidirectional exchange between multiple agents. A computational system will be developed to intelligently decide how to distribute SA. Together with other colleagues in the Future Everyday group of Industrial Design, you will explore how multiple AVs can communicate with multiple other road users (includingVRUs) and investigate how AVs can ensure that the conveyed information is compatible with the behaviour of other road users.
You will develop further the coupled simulator for human factors research (https://github.com/bazilinskyy/coupled-sim
). With help of the simulator, the candidate will conduct experiments, where multiple participants will meet and interact in the same virtual world. Through simulator and on-road experiments, the candidate will investigate how AVs can assess road users' SA from movements and posture, communicated through connected portable sensors (equipped on AVs, cars and VRUs) and explore mechanisms of assessing SA.
The results of the project will be used by automotive manufacturers and governments to coordinate the actions of future traffic participants. There will be close interaction with other researchers at the Department of Industrial Design and with the industry. There will be opportunities to conduct joint experiments with other PhD candidates. The candidate will work very closely with Dr. Bazilinskyy, who has a background in Computer Science and Human Factors. It is a challenging technical project that promises to revolutionise the field of Human Factors of automated driving. If you're an ambitious person who's not afraid of tackling a complicated challenge, this is the project for you!