Employing Nelder-Mead and Gravitational Search Algorithm Methods in Point Clouds Registration Process

Abstract

Abstract In reverse engineering it may be required to perform multiple digitizing due to the size and part complexity. Registration process for point clouds is to find the geometric transform between them in which all point clouds are transformed into a single absolute coordinate system. The process of registration is strongly affected by inaccurate data and may fail in the case of noisy data, hence other methods are usually sought to find the transformation matrix. This paper tries to solve the problem in practical applications. The Nelder-Mead and Gravitational Search Algorithm methods were employing for point clouds registration for the first time. The registration was also performed using Singular value decomposition and Genetic Algorithm methods. The four methods were compared in terms of convergence, accuracy and computation time.