Maneuver-based Montion Planning for UAV in both Static and Dynamic Environments

This paper studies path planning for the UAV based on a hybrid framework, which can be obtained by using a maneuver automaton to describe the relationship of trim states and maneuver states of the UAV. For methods based on kinetic equations, it may be difficult to obtain an optimal path satisfying dynamics constraints such as flight envelope restrictions. To address this difficulty, this paper proposes a maneuver-based scheme, where a motion library based on dynamics equations of the UAV is essential and information about the environment is embedded into an occupancy-based map, the planning scheme is modeled as a multi-objective program with constraints, the feasibility of which is demonstrated by analyzing path planning problems both in static and dynamic environments. In addition, a modified genetic algorithm (GA) is adopted to improve the convergence speed of the multi-objective program to obtain a primitive sequence. Simulation results illustrate the superior performance of the proposed scheme.