Abstract/Summary

Driverless cars are emerging slowly but bear the opportunity to improve the traffic system efficiency and user comfort. For the near future, a mix of human-driven and driver-less vehicle co-existence is inevitable. The quest to integrate autonomous and human-driven vehicles has created numerous critical questions in the road traffic system: Should humans remain to navigate the wheel? How can autonomous and human-driven vehicles co-exist efficiently? What are the prospects of breaching traffic data security? How can we address some social driving problems ranging from congestion reduction, communication among traffics, and many more? To a large extent, efficient control and supervision of mix-traffic behaviours at the road intersection will go a long way to ameliorate the concerns envisaged in the autonomous vehicle integration process. Mixed traffic co-existence is associated with lateral and longitudinal direction (2D) behaviour which needs communication among vehicles. These stated mixed-traffic characteristics principle contradicts the car-following model, which only describes longitudinal vehicle interaction in homogeneous traffic. A model predictive control-based node reservation technique is developed to optimise the flow of mixed vehicles at a discrete-time step based on the human-driven vehicles (HVs’) estimated driving behaviour. The main contributions of this thesis are employing the existing 1-dimensional homogeneous car-following model strategies to a 2-dimensional heterogeneous traffic system, synchronising the two-vehicle type control communication component (human perception for HVs, and vehicle to X (vehicle and or infrastructure) for autonomous vehicles (AVs)), modelling driving behaviour with vehicle-type-contingency, and a balanced impact assessment of AVs in a mixed-traffic scenario to serve as an integration pattern. To quantify the benefit of the proposed Node Reservation strategy, a simulator is developed, and three traffic management strategies are integrated: traffic light control method, collision avoidance with safe distance method, and the node reservation method. The proposed simulation model is validated, and experiments are conducted with varying traffic intersection control strategies and vehicle type proportions. The obtained results demonstrate that the node reservation strategy has a high throughput with minimal delay and braking.