Precise single-axis laser scan robot calibration method on basis of ball fitting and system

Abstract

The invention provides a precise single-axis laser scan robot calibration method on the basis of ball fitting and a system. The method comprises steps that 1, the position of a non-mirror target ball is made to be basically at the middle position of a single-axis robot stroke; 2, a line laser sensor is driven through motion of a single-axis robot to scan the non-mirror target ball to acquire ball surface point cloud; 3, the ball surface point cloud and miscellaneous points which are repeatedly scanned are removed, noise point cloud not belonging to a ball surface is eliminated; 4, rotation transformation for the ball surface point cloud is carried out, re-fitting of the ball surface point cloud after transformation is carried out, and a fitting residual error is acquired; and 5, whether the fitting residual error satisfies an iteration termination condition is determined, if yes, an optimal solution is acquired; if not, the process returns to the step 4. The method is advantaged in that dependence on an external precise apparatus is not needed, strong practicality is realized, and the method is applicable to industrial field applications in multiple occasions.

Description

technical field [0001] The invention relates to the field of robot three-dimensional laser scanning detection, in particular to a method and system for precise calibration of a single-axis laser scanning robot based on ball fitting. Background technique [0002] The line laser sensor mainly uses the measurement principle of space triangulation, the contour curve of the measured surface is collected by the CCD, and the three-dimensional geometric point cloud data of the surface of the object is obtained through computer processing. Three-dimensional laser scanning technology refers to the use of robot motion to drive the line laser sensor to continuously scan the outline of the object. This technology can be used to quickly collect complete three-dimensional geometric point cloud data on the surface of the measured object, and truly describe the overall structure and morphological characteristics of the measured object. Generate a 3D data model and combine the 3D laser scanni...

AI Technical Summary

Problems solved by technology

Therefore, if the line laser sensor is directly installed on the industrial robot to realize three-dimensional scanning, the obtained point cloud has low precision and the effect is not good.

Images