中国机械工程 ›› 2014, Vol. 25 ›› Issue (6): 766-770.

• 智能制造 • 上一篇    下一篇

基于支持向量机的点云曲面刀具路径规划

吴福忠   

  1. 绍兴文理学院,绍兴,312000
  • 出版日期:2014-03-26 发布日期:2014-04-11

Tool Path Planning for Point Cloud Surface Based on LS-SVM

Wu Fuzhong   

  1. Shaoxing University,Shaoxing,Zhejiang,312000
  • Online:2014-03-26 Published:2014-04-11

摘要:

针对单值散乱点云曲面刀具路径规划问题,提出了一种基于最小二乘支持向量机的计算方法。在计算过程中,将点云数据向平面投影,得到二维点集。应用网格划分和边界网格内测量点高斯映射技术,提取平面区域内的边界特征点。用边界特征点定义点云曲面的实际加工区域,在此区域内规划平行等间距刀具路径。应用最小二乘支持向量机拟合点云数据,求得被加工曲面的连续表达模型,经此模型将二维刀具路径数据向三维空间映射,求出刀触点数据。将刀触点经法向偏置计算,求得刀位点。实例验证证明,该方法能较好地解决信息不完备散乱点云曲面刀具路径生成问题。 

关键词: 最小二乘支持向量机, 点云曲面, 刀具路径, 数控加工

Abstract:

Aiming at the tool path planning problem of single-valued scattered point cloud surface, an algorithm based on LS-SVM was proposed. In the calculation process, a 2D point set was acquired by projecting the measured points onto plane, and the boundary feature points were extracted by method of meshing and Gauss mapping of points within boundary mesh. Thus the actual machining region was defined by the boundary points, and the parallel iso-planar tool path was planned in the planar region. Then the fitting surface model of the measured points was constructed by using LS-SVM. Mapping 2D tool paths to 3D space with this model, the cutter contact points were found out. Ultimately the cutter location points were obtained by offsetting cutter contact points along normal direction. The proposed algorithm was verified by an example. The results indicate that the present method can better solve the tool path generation problem of the incomplete measured points.

Key words: least square support vector machine(LS-SVM);point cloud surface, tool path, NC machining

中图分类号: