【摘要】 针对多点成形件成形误差检测中的三维数据配准问题 ,提出了一种用两步法将 CAD模型数据和三维测量数据点全局配准的方法。利用遗传算法具有全局搜索能力的特点 ,以四元数法中的 3个参量作为优化解空间 ,采用自适应控制优化参数的方法 ,达到 CAD模型数据和三维测量数据点的全局粗配准。以粗配准的结果作为初始值 ,用 ICP算法修正误差以达到精确配准。采用两步法配准克服了标准 ICP算法难以解决的局部最小问题 ,本算法也可广泛应用于精密测量时测量结果的比较分析。以马鞍形曲面制品为例 ,给出了配准结果。更多还原

【Abstract】 In this paper,a new method for the global registration of the CAD modeling data to the 3D measured data is presented.It solves the problem of registration of 3D data prior to shape error inspection of multi-point forming part.Three parameters are applied of unit quaternions method as optimizing space, and global rough registration between the CAD modeling data and the 3D measured data is obtained.ICP algorithm in which the results of rough registration are used as initial value is processed for adjusting the error and to improve the accuracy of registration of two point sets. The method solves the problem of local minima of ICP algorithm,and the described method is also useful in the field of comparative analysis for precision measurement.Taking a saddle surface part as an example ,some results are discussed.更多还原