MILP based autonomous vehicle path-planning controller for unknown environments with dynamic obstacles

Autonomous vehicles (AVs) manoeuvring in unknown environment require path-planning algorithms that are safe, yet optimal to circumvent dynamic obstacles with minimum fuel-cost. This investigation presents an autonomous vehicle path-planning (AVPP) controller that uses mixed integer linear programming to decide the blending and switching actions among possible vehicle behaviours depending on local sensed information. Our results illustrate the safety and optimality of the controller for AVPP in unknown environments with dynamic obstacles. Comparison with existing methods shows that the proposed method is more robust to collisions than the fuzzy and extended Kalman filter based arbitration mechanism studied in literature. Further, as behaviours breakdown the complex path-planning problem into simple tasks, controller realisation becomes simple. © 2016 Inderscience Enterprises Ltd.