Autonomous Vehicle Mission Planning

Autonomous vehicle mission planning is one of the core algorithms for the autonomy of vehicles. Path planning is part of mission planning algorithms providing the desired path commands to achieve given missions. We study two widely used path planning algorithms: the potential field method and the graph-based method. A potential function method uses fluid dynamics modelling for a sink and a source to generate a fluid flow. A graph-based method is a sampling approach. It discretizes the spatial space and constructs a finite number of paths between them. It is a flexible and powerful method to solve the path planning problem in complex obstacle environments. We investigate a target-tracking problem, which is a frequent mission planning problem. The continuous dynamics are discretized, and a worst-case algorithm solves the optimal target-tracking control problem. Detailed derivations of the optimization problem and the solution are presented. The performance of the algorithm is demonstrated using simulations.