Seminar held by Prof.Dong Ngoduy (2018 July 31th)
Prof.Dong Ngoduy from University of Canterbury, New Zealand held a seminar.
This semnar is co-hosted with Kyoto University Geoinformatics Lab.(Uno Lab.) and Ritsumeikan University Intelligent Transportation Systems Laboratory(Shiomi Lab.).
Lecturer: Prof.Dong Ngoduy; Chair, Connected Traffic Systems Lab, University of Canterbury
Title: Modelling heterogeneous platoon dynamics under the connected environment
Abstract:This research aims at developing an advanced model based control framework which can efficiently integrate practical vehicular communication for conventional and connected and autonomous (CA) vehicles in the heterogeneous intelligent traffic systems. In general, a CA vehicle can obtain neighbouring information via vehicle-vehicle (V2V) communication and/or vehicle-infrastructure (V2I) communication (hereafter, V2X communication for short), and then adopt a suitable control law to achieve certain objective, such as maintaining a constant inter-vehicle spacing within vehicles or smooth driving patterns. These CA vehicles will fundamentally transform the conventional (human driven) transportation as well as global economy and society. Accordingly, the critical issue of integrating V2X communications into transport system has attracted great concerns in a number of research projects. Nevertheless, a fully deploying CA cars on roads still remains a long way to go. It has been predicted that by 2030 CA vehicles will make up a significant share of the vehicle market, which indicates a long lifespan for traffic flow consisting of both conventional and CA vehicles. Specifically, due to the long-term evolution of automated vehicle technologies, in this research, we refer to the CA vehicles as the ones with V2X communication capability and different level of automation. Modelling and managing the dynamics of such heterogeneous traffic flow is a very challenging task due to the complex (asymmetric) interactions between conventional and CA vehicles. This research explores the idea of combining the theory of traffic flow and ICT to develop a model-based traffic prediction and control framework for future transportation systems. Our research seeks to improve the utilisation of existing capacity by (i) using improved traffic prediction models (ii) better communication protocols, (iii) providing better information to users, and iv) better traffic control strategies (i.e. ramp metering and speed limits).