MPC-BASED AUTONOMOUS DRIVING CONTROL WITH LOCALIZED PATH PLANNING FOR OBSTACLE AVOIDANCE AND NAVIGATING SIGNALIZED INTERSECTIONS

Connected and autonomous vehicles are becoming the major focus of research for the industry and academia in the automotive field. Many companies and research groups have demonstrated the advantages and the requirement of such technology to improve the energy efficiency of vehicles, decrease the number of crash and road accidents, and control emissions. This research delves into improving the autonomy of self-driving vehicles by implementing localized path planning algorithms to introduce motion control for obstacle avoidance during uncertainties. Lateral path planning is implemented using the A* algorithm combined with piecewise Bezier curve generation which provides an optimum trajectory reference to avoid a collision. Model Predictive Control (MPC) is used to implement longitudinal and lateral control of the vehicle. The data from vehicle-to-everything (V2X) communication infrastructure is used to navigate through multiple signalized intersections. Furthermore, a new method of developing Advanced Driver Assistance Systems (ADAS) algorithms and vehicle controllers using Model-In-the-Loop (MIL) testing is explored with the use of PreScan®. With PreScan®, various traffic scenarios are modeled and the sensor data are simulated by using physics-based sensor models, which are fed to the controller for data processing and motion planning. Obstacle detection and collision avoidance are demonstrated using the presented MPC controller. The results of the proposed controller and the scope of the future work conclude the research