A Pose Estimation System and Method for Complex Parts Based on 3D Measurement Point Cloud

Abstract

The invention belongs to the field of automatic measurement, and specifically discloses a complex part pose estimation system and method based on three-dimensional measurement point cloud, which realizes the complex part pose estimation through the following steps: calibration of the six-degree-of-freedom industrial robot terminal flange and The grating area scanner measures the transformation matrix between the coordinate systems; the six-degree-of-freedom industrial robot drives the grating area scanner to scan the workpiece to obtain the three-dimensional point cloud data of the workpiece; it will be The 3D point cloud data of the measured workpiece is converted to the robot base coordinate system to obtain the converted 3D point cloud data; the converted 3D point cloud data is matched with the 3D design model of the measured workpiece to obtain the measured workpiece relative to the robot. The pose of the base coordinate system is used to complete the estimation of the pose of complex parts. The invention has a wide measurement range, can realize accurate measurement of multi-regions and complete shape of the workpiece to be measured, and can accurately obtain the pose of the workpiece to be measured.

Description

technical field [0001] The invention belongs to the field of automatic measurement, and more specifically relates to a system and method for estimating the pose of a complex part based on a three-dimensional measurement point cloud. Background technique [0002] Robot 3D automatic measurement needs to plan a complete and collision-free path, and the path planning must know the attitude of the workpiece relative to the robot's base coordinate system. The traditional workpiece pose estimation method includes manual alignment method, which relies on manual operation, is random and inaccurate; there is also a method of using visual positioning to obtain the pose of the workpiece, which uses the characteristics of the workpiece to perform image matching to obtain Workpiece pose, this method is not suitable for workpieces with inconspicuous features, and the calculation is complex. [0003] These traditional measurement methods are not only complicated, but also difficult to obta...

AI Technical Summary

Problems solved by technology

The traditional workpiece pose estimation method includes manual alignment method, which relies on manual operation, is random and inaccurate; there is also the method of using visual positioning to obtain the pose of the workpiece, which uses the characteristics of the workpiece to perform image matching to obtain Workpiece pose, this method is not suitable for workpieces with inconspicuous features, and the calculation is complicated

[0003] These traditional measurement methods are not only complicated, but also difficult to obtain accurate workpiece poses, resulting in deviations in subsequent path planning, especially when the workpiece is very large, it is very difficult to obtain the workpiece coordinate system, which will lead to subsequent robots. There are large deviations in the planning of measuring points, which may lead to unexpected collisions between the robot measuring system and the external environment

Images