Multi-structured light binocular composite vision weld joint tracking method and device

Abstract

The invention relates to a method and a device for tracking a multi-structure optical binocular composite vision weld joint, and relates to the technical field of automatic welding, in particular to the technical field of vision sensing of a welding robot. The device comprises a multi-line laser generator (10) for projecting a structured light stripe to a weld joint region, a vidicon A(3) which is provided with a narrow bandpass filter (1) and a neutral light reducing plate A(2) and is used for collecting a weld joint character laser stripe image, a vidicon B(7) which is provided with a bandpass filter (8) and a neutral light reducing plate B(9) and is used for collecting an image not including laser stripe, and a controller (11) connected with the vidicon A(3) and the vidicon B(7). The method integrates image information of two vidicons, and calculates the current perfect position of a welding gun so as to obtain differential value between the perfect position and the actual position; the controller converts the differential value into an analog signal or wireless communication to output; and the welding gun is controlled to correct deflection of an electromotor, so the aim of automatic tracking of the weld joint is realized.

Description

technical field [0001] The invention relates to a multi-structured light binocular composite vision welding seam tracking method and device, and relates to the technical field of automatic welding, in particular to the technical field of welding robot vision sensing. technical background [0002] At present, there are many kinds of sensing methods applied to welding robots. Among these sensing methods, the visual sensing method has the advantages of large amount of information, non-contact, fast, high precision and large detection range. Workers love it, and it has become a research hotspot in robot welding sensors. At present, the relatively mature research at home and abroad is the visual sensing method using laser or structured light. This sensing method is accurate in measurement and strong in anti-interference ability, but there is a disadvantage that there is a certain distance between the measurement point and the welding point. Leading error is easy to occur when cu...

AI Technical Summary

Problems solved by technology

Lu Minghao et al. published in "Mechanical Design and Manufacturing" (2008, No.7: 123-125) "Study on the Seam Tracking System of Spiral Steel Pipe Submerged Arc Welding Based on Structured Light" using multi-structured light technology for welding seam tracking. Tracking, but this technology has no essential difference from the single structured light technology, and still cannot eliminate the leading error generated when tracking the curved weld seam

Although the vision sensor introduced by Zhou Lv and others in the patent "Welding Robot Servo Binocular Vision Sensor" (Application No. 200410067328.1) uses binocular vision, it actually only uses one of them for seam tracking. The camera is illuminated by natural light or arc light, which has low precision and poor anti-interference ability

Images