A Method of Obstacle Recognition for Laser Type Complicated Trajectory Welding Seam
A laser-based technology for obstacle identification, applied in welding equipment, arc welding equipment, manufacturing tools, etc., can solve the problems of low accuracy and stability
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0013] Embodiment 1, the scanning laser-based fillet weld obstacle recognition method, the obstacle recognition system consists of 1-adaptive swing amplitude change execution module, 2-laser signal acquisition and processing module, 3-obstacle recognition controller , 4-laser vision transceiver module, 5-adaptive swing amplitude calculation and drive module, 6-traveling mechanism, in which the laser obstacle recognition sensor is composed of 2, 3, 4 modules, and the adaptive swing amplitude swing device is composed of 1 , 5 modules; the laser vision transceiver module 4 is driven by the self-adaptive swing and variable amplitude execution module 1, and reciprocates at a certain frequency perpendicular to the complex track weld 9. When the welding direction is moving at a constant speed, the laser vision transceiver module receives the sampled eigenvalue signal reflected from the laser working position point 7, and the laser signal acquisition and processing module 2 extracts it...
Embodiment 2
[0014] Embodiment 2, the specific implementation of the laser-type obstacle recognition sensor and the multivariate statistical discrimination algorithm is: the laser vision transceiver module receives the sampled feature value signal reflected from the laser working position point, and extracts its characteristic value signal through the laser signal acquisition and processing module. The position peak and the obstacle recognition controller will sequentially form the first p peak values into a p-dimensional eigenvalue matrix, which is recorded as the overall X (0) , and then each new p / n position peak value replaces the first p / n position peak value to form a new p-dimensional position matrix, which is recorded as the overall X (1) . By analogy, a multivariate same-dimensional dynamic matrix X is formed (i) ,(i≥0). According to the output characteristics of the laser vision sensing device used, it can be seen that the p-dimensional position matrix formed by the peak valu...
Embodiment 3
[0029] Embodiment 3, the self-adaptive luffing oscillator is used to adjust the signal acquisition range of the laser vision signal acquisition device, so as to ensure the adaptability of the laser on the complex track workpiece. It is characterized in that the device is composed of a swing execution module, a swing adjustment control calculation module and a drive module. According to the output characteristics of the obstacle recognition sensor, the fitting function relationship l=f(x ij ), where l is the actual length of the optical path, x ij is the peak value of any position; moreover, the multiple feedback signals analyzed by the multivariate statistical discriminant algorithm in the previous embodiment include the mean value vector and expected value μ i and so on; mu i =h(x ij ), the constructor but refer to Figure 4 , when the weld trajectory of the workpiece changes, p=4 is taken, and the peak values at the four positions corresponding to Figures a, b,...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More