Development of a framework for sensor- and communication-assisted Vehicle Dynamic.

The purpose of this thesis is to create a co-simulation framework composed of ns3 module for simulating wireless communication network; CarMaker module for simulating vehicle dynamics, the vehicle on-board sensors, and the vehicle traffic; Python engine in control of data collection and control strategies; and MATLAB/Simulink module as an interface between the Python engine and CarMaker. The proposed framework uses Vehicle to Infrastructure communication model for creating the Cooperative Adaptive Cruise Control(CACC) system. The network infrastructure node periodically receives Cooperative Awareness Messages (CAM) from connected vehicles. The packets contain data about each vehicle position, heading, acceleration, velocity. Then the infrastructure node will run a trajectory-based collision avoidance algorithm which forecasts the possible collision between vehicles. When a collision is detected, the node will send the updated acceleration to the Ego car simulated in CarMaker. By this way, we can analyse the performance of ADAS and Communication models.We have tested the framework against multiple real-world traffic scenarios with two different cases: Vehicles with ADAS only and Vehicles with ADAS and OBU (connected vehicles). A vehicle only with ADAS will fail to detect the vehicles falls under its blind spot whereas a vehicle with radar and OBU will be able to detect every vehicle in its surroundings and adjust its acceleration to avoid the collision. For decision making, the framework can switch between ADAS sensor data and data received through Communication models based on its priority. The results proved that an integrated framework would be a valid starting point to investigate the impact of connected vehicles.