Application of iterative closest point algorithm in automatic flight of speedy UAV

In order to realize the autonomous location and attitude estimation for speedy UAV, this paper designs an improved Iterative Closest Points (ICP) algorithm for fixed-wing UAV with a laser range finder mounted onboard. This improved ICP adopts a K-d tree and AK-d tree mixed closest point searching strategy to enhance its rapidity, and an automatic trim approach to drop out outlier points to improve its reliability. In order to check the feasibility of the proposed algorithm, flight experiment is conducted. Experiment result demonstrates the good performance of the algorithm.