A virtual-waypoint based artificial potential field method for UAV path planning

Abstract

Unmanned aerial vehicle (UAV) path planning is an important preliminary step in UAV mission planning. Artificial potential field (APF) UAV path planning method is often employed to build an intuitive model which is updated by the additional forces pushing the trajectory away from threats of non-cooperative objects and toward the predefined target. However, APF has its own weaknesses, one of which is the local minima close to concave obstacles. To tackle situation mentioned above, we propose a novel additional force model based on virtual waypoints. This model provides additional forces for candidate planning points falling into local minimum area and help them escape from local minimum area. Once the candidate is trapped in minimum area, firstly we calculate measure factor according to the planning area containing current candidate and the target point, and thus determine position of the virtual waypoint. Then the virtual waypoint generates additional control force so that the candidate can run away from the local minimum area. Finally, the simulation results show that the method can solve the local minimum problem in planning space.