Simultaneous Localization and Mapping (SLAM)

Abstract

There are many algorithms for simultaneous localization and mapping (SLAM) such as the extended Kalman filter (EKF) algorithm ,Rao-Blackwellized particle filter (RBPF) and Iterative Closest Point(ICP) algorithm. This paper choose ICP algorithm for SLAM, because ICP algorithm is easier to implementation and is need less memory for central processing unit. The Iterative closest point (ICP) is originated in [6]-[8]. This algorithm is often used in computer vision and pattern recognition. It can find a spatial transformation to match two point sets, and reconstruct 3D surfaces. The key idea of this algorithm is to calculate the transformation matrix T, which includes rotation R and translation t , between a point cloud P P Pp , p , … , p and another point cloud Q Q q , q , … , q . The transformation T is defined as equation (1). 0 1 (1)