Research on Obstacles Avoidance for UAV SLAM

Abstract

Abstract: When UAV is implementing the simultaneous localization and mapping (SLAM) problem, the environment where UAV is flying exists unavoidable solid or moving obstacles because of its unknown character, which threatens the flying safety and the completeness of SLAM mission. To conquer this problem, an improved artificial potential field algorithm is proposed to simultaneously accomplish obstacle avoidance of UAV and SLAM mission based on a potential field function containing the distance from UAV to the goal and from UAV to the obstacles and the covariance of features. This algorithm is simulated and tested based on the built UAV plane motion model. The result shows that the proposed algorithm is effective to avoid the obstacles while implementing SLAM for UAV.