NURBS Curve Fitting Based on ACO Algorithm for Solving Optimal Control Vertices

• Published in: 2024 International Conference on Artificial Intelligence and Power Systems (AIPS) pp. 197 - 202
• Main Authors: Miao, Liqin, Liao, Chaoneng, Zhang, Dashun, Liang, Huaidan, Wang, Jing
• Format: Conference Proceeding
• Language: English
• Published: IEEE 19.04.2024
• Subjects: ACO algorithm; Curve fitting; Feature extraction; Fitting; Least mean squares methods; NURBS curve fitting; Optimal control; Power systems; Splines (mathematics); Surface reconstruction; Surface topography; Tool path optimization
• DOI: 10.1109/AIPS64124.2024.00049

A NURBS curve fitting algorithm based on ACO algorithm is proposed for tool path generation, which takes the sum of the squares of the deviations between the tool points and the fitted curve as the objective function. Firstly, discretizes the data points by using the isoparametric method. Secondly, the curvature extrema and inverse curvature points are found out through the adaptive feature point extraction method. And then the initial control vertices are back-calculated through the least squares method. Finally, the ACO algorithm is used to find out the The optimal control vertices are finally found by the ant colony algorithm, thus realizing a better tool path fitting for NURBS curves. And through the experiment, for the algorithm proposed in this paper and the least squares method, with the average error and the mean square deviation as the judgment standard. It is concluded that the algorithm in this paper improves the fitting effect by 50.8% compared with the least squares method, which verifies the superiority of this paper's algorithm.