Binocular-vision-based method for monitoring and tracking pose of moving platform of parallel mechanism

Abstract

The invention discloses a binocular-vision-based method for monitoring and tracking the pose of a moving platform of a parallel mechanism. Moving-platform circular identification area pose images shotby a left camera and a right camera simultaneously at different angles are collected; with image recognition and feature extraction techniques, an edge search model is constructed; limited multiple points of the edge of the moving platform are extracted; on the basis of a least square method and a meta heuristic intelligent algorithm, ellipse fitting is carried out on edge points of a moving platform identification area to obtain a central pixel coordinate value of the moving platform identification area; on the basis of a camera calibration toolbox and a calibration plate with the known size, parameter calibration of a monocular camera is performed and a binocular stereo matching algorithm is obtained to realize dual-target setting; and on the basis of a camera imaging principle relationship, internal and external parameters of the left and right cameras and a relationship between the central pixel coordinate of the moving platform identification area and the world coordinate are deduced, and a real-time three-dimensional coordinate of the central point of the moving platform identification area is reconstructed to realize real-time monitoring and tracking of the pose of the moving platform. The binocular-vision-based method has advantages of high measurement accuracy and reduced error and the like.

Description

technical field [0001] The invention relates to the field of pose monitoring and tracking, in particular to a binocular vision parallel mechanism motion platform pose monitoring and tracking method. Background technique [0002] The parallel mechanism consists of a fixed platform, a moving platform and multiple intermediate branch chains, and has the advantages of high specific rigidity, compact structure, strong bearing capacity, and good stability. However, the movement of the moving platform of the parallel mechanism is jointly determined by the movement of multiple intermediate branches, the coupling degree is high, and the control is relatively complicated. Accurately measuring the pose of the moving platform at the output end is of great significance to the overall high-performance control of the parallel mechanism. Traditional measurement methods include contact measurement, monocular measurement, binocular single-point measurement, etc. Among them, contact measuremen...

AI Technical Summary

Problems solved by technology

However, the movement of the moving platform of the parallel mechanism is jointly determined by the movement of multiple intermediate branches, the coupling degree is high, and the control is relatively complicated. Accurately measuring the pose of the moving platform at the output end is of great significance to the overall high-performance control of the parallel mechanism

Traditional measurement methods include contact measurement, monocular measurement, binocular single-point measurement, etc. Among them, contact measurement requires the measurement equipment to follow the movement of the mechanism, which is greatly affected by the movement. At the same time, the measurement equipment will inevitably change the mass distribution of the mechanism and affect The normal operation of the mechanism; the monocular measurement uses one camera to identify and track the movement of the mechanism, the uncertainty in the direction measurement is large, and the depth recovery is also difficult; the binocular single-point measurement is based on two cameras simultaneously identifying the movement of a single mark point on the mechanism , limited by the environment, it is easy to lose information and affect the tracking effect

Images