Constrained UAV mission planning: A comparison of approaches

In this paper the optimal timing of air-to-ground tasks is considered. A scenario is examined, where multiple unmanned air vehicles (UAVs) must perform one or more tasks on a set of geographically dispersed targets in the presence of no-fly zones. Strict inter-task timing constraints apply, which have been chosen by BAE Systems' Military Air & Information (MAI) business to reflect current UAV usage. The optimal assignment of these tasks requires cooperation amongst the vehicles in order to generate a plan that is efficient, with respect to overall mission duration and satisfies all problem constraints. A number of previous papers have compared the benefits of employing numerous optimisation algorithms to this class of problem. Recently, auction techniques have been investigated for coordination of multiple autonomous assets and this paper compares two auctioning algorithms, a meta-heuristic algorithm and a mathematical programming approach for solving such constrained task assignment problems.