Rapid Multi-Query Path Planning For A Vertical Take-Off and Landing Unmanned Aerial Vehicle

Kurzfassung

This paper presents an online approach for multi-query path planning that enables an unmanned aircraft to quickly react to newly located or changed obstacles. All three spatial dimensions are considered during planning, as it is likely to be necessary for urban terrain with overhangs or inside buildings.Asampling-based motion planning approach is combined with an anytime path search such that an online trade-off between path optimality and rapid path computation is possible. The presented global path planner can rapidly handle global cost changes acquired during flight. Moreover, the multi-query property of the planning system supports combinatorial optimization problems such that the decisional autonomy is maximized during flight. This increase in system autonomy is achieved for holonomic maneuvering through scenes with unforeseen obstacle changes. This way, a common constraining requirement for unmanned aircraft can be relaxed: with onboard sensing and planning capabilities, a highly reliable, real-time remote control link is not necessary to ensure collision-free flight and higher level decision-making.