Vehicle automation and safety
In this research theme we focus on the interaction of different road users with automation. We are investigating for example, how pedestrians and cyclists adapt their behaviour when interacting with automated vehicles and automated shuttles? Similarly, how would human drivers change and adapt their driving behaviour when interacting with connected and automated vehicles in mixed traffic? and What are the impacts of authority transitions on traffic safety?
Relevant research:
Narayana Raju: Implications of integrating CAVs in mixed traffic on the traffic flow efficiency and safety
Implementing the new knowledge on humans’ behavioral adaptation when interacting with AVs, and AVs behavioral models in an existing open-source simulation platform. Followed by assessing the implications of different scenarios on traffic flow efficiency and safety, and consequently propose recommendations regarding the infrastructure (physical and digital) requirements.
Solmaz Razmi Rad: Performance and safety evaluation of dedicated lanes for connected and automated vehicles
Considering the transition period when both Connected and/or Automated Vehicles (C/AVs) and Manual Vehicles (MVs) will be present on our road network, it is crucial to understand how they interact in a mixed environment when CAVs are clustered in platoons, and whether the behaviour of human drivers is influenced by CAV system or platoons.
Yongqi Dong: Data-driven research for expanding AVs’ operational design domain in mixed traffic
State-of-the-art deep reinforcement learning models are being explored to develop reliable models for AVs’ driving policies under selected maneuvers involving both longitudinal and lateral control. Metrics regarding safety (e.g., Time To Collision, Encroachment Time, Mean Lane Position), efficiency (e.g., total travel time , average traffic speed and density ), and social compliance ( e.g., Social Value Orientation, Social Norm) will be integrated to form the proper loss function for DRL. The DRL model will be trained and validated on an open-sourced traffic simulation (microscopic) platform.
Siri Hegna Berge: Human-machine interface (HMI) on bicycles promoting transparent AV interactions
Cyclists are expected to interact with AVs in future traffic, yet we know little about the nature of this interaction and the safety implications of AVs on cyclists. On-bike HMIs and connecting cyclists to AVs and the road infrastructure may have the potential to enhance the safety of cyclists. In this project, we will focus on understanding the factors that constitute cyclist-AV interaction and investigate the efficiency and acceptance of on-bike HMIs and connected bicycles among cyclists.
Nagarjun Reddy: Studying human drivers’ behaviour in mixed (HDV and AV) traffic environments
The rapid deployment of automated vehicles (AVs) is certain to bring a change in the experience of human drivers, who would still be significantly present on roads along with AVs, at least in the near future. This could lead to changes in driving behaviour of HDVs, also known as behavioural adaptation. Such behavioural adaptation at the tactical and operational level would naturally have an effect on traffic safety and traffic flow. Therefore, studying and understanding HDVs’ behavioural adaptation will be critical to foresee the state of future traffic. Investigation into the factors that affect human driving behaviour and the precise nature of their effect will be studied.