Dynamic matrix and model predictive control for a semi-auto pilot car

Dynamic matrix and model predictive control in a car aims at vehicle localization in order to avoid collisions by providing computational control for driver assistance which prevents car crashes by taking control of the car away from the driver on incidences of driver's negligence or distraction. This paper provides ways in which the vehicle's position with reference to the surrounding objects and the vehicle's dynamic movement parameters are synchronized and stored in dynamic matrices with samples at regular instants and hence predict the behavior of the car's surrounding to provide the drivers and the passengers with a driving experience that eliminates any reflex braking or steering reactions and tedious driving in traffic conditions or at junctions.It aims at taking corrective action based on the feedback available from the closed loop system which is recursively accessed by the central controller of the car and it controls the propulsion and steering and provides a greater restoring force to move the vehicle to a safer region.Our work is towards the development of an application for the DSRC framework (Dedicated Short Range Communication for Inter-Vehicular Communication) by US Department of Traffic (DoT) and DARPA (Defense Advanced Research Projects Agency) and European Commission- funded Project SAVE-U (Sensors and System Architecture for Vulnerable road Users Protection) and is a step towards Intelligent Transportation Systems such as Autonomous Unmanned Ground and Aerial Vehicular systems.