Particle Filter based Position Arrangement Method for Point Cloud Data Registration

Abstract

In this paper, we propose a position arrangement method for the point cloud data registration. Typically, the iterative closest point method performs well for the rigid registration of the point cloud data. However, the iterative closest point method failed the registration on an abnormal initial position. Therefore, we propose the particle filter based method for the position arrangement. The proposed method consists of the 3 steps: particle draw, particle evaluation, and update. In order to enhance the accuracy, we control the range of particle. And we control the sampling rate to improve the speed of proposed method. Experimental results show that the propose method successively register on various initial position.