Cooperative Traffic Control Solution for Vehicle Transition from Autonomous to Manual Mode exploiting Cellular Vehicle-to-Everything (C-V2X) Technology

Nowadays, automated vehicles represent a promising technology to face the stringent requirements for safety and traffic efficiency in the automotive environment. Driving responsibilities will be gradually addressed to the machine, and the role of human pilots will be progressively reduced to passengers. The interaction between passengers and the automated system will create different risks that have not been considered in the past. In particular, the transition between autonomous and manual mode is understood as a risky situation. During the transition, the driver manifests driving irregularities and loss of situation awareness that may endanger himself and other participants on the road. Hence, the vehicle transitioning needs a higher quantity of space around it to be considered safe. However, no effective solution has been developed yet. This thesis aims to design a cooperative traffic control solution that will manage the movements of the group of vehicles to increase the free space around the one transitioning. It will exploit another tool that will play a fundamental role in the future of the automotive industry: connected vehicles technology. C-V2X technology will create a medium for vehicles to exchange information and cooperate. A controller managing the cooperation between vehicles has been developed to help a smooth and safe vehicle repositioning. The controller will be positioned in a centralized computing facility and it will communicate with all the vehicles. The controller defines rules to move vehicles together and enlarge the free space around the vehicle transitioning without collisions. The rules are modeled by a spring-mass-damper system, that can be exploited to control the longitudinal behavior of automated vehicles. In particular, the spring-mass-damper system can manage smooth migration between vehicle dispositions without oscillations. A computer simulation is used to test the performance of the proposed traffic control system. The simulation environment is constituted by three main components: traffic flow, controller and communication network. It has been tested with the software VEINS, which provides interaction between a network simulator (OMNeT++) and a traffic simulator (SUMO). The traffic flow represents the interactions between vehicles. The controller analyzes the data and sends control messages to all vehicles. The communication network will share the data concerning vehicles’ position and speed and control messages. The proposed cooperative vehicle control system demonstrated to reduce the risks of the transition with the smooth motion of vehicles. The controller is able to achieve the safety requirements without reducing the level of comfortability of vehicles’ passengers.