Robot space attitude resolving method based on structured light array remote identification

Abstract

The invention discloses a robot space attitude resolving method based on structured light array remote recognition, which comprises the following steps: S1, recognizing structured light spots, and acquiring different light spots on a robot; S2, analyzing light spots, and calibrating the posture of the robot according to geometrical characteristics formed by different light spots; S3, repositioning the mass center of the light spots; S4, measuring and calculating the relative attitude; and S5, resolving a space attitude. According to the method, the structured light array on the surface of the robot is remotely recognized through a pan-tilt camera, under the condition that the internal reference of the camera is fixed, the space attitude of the robot is calculated according to the geometrical parameters of the structured light and the external reference of the pan-tilt camera, and auxiliary correction can be conducted on an IMU measurement value. The method avoids the problem that the remote imaging of the camera is blurred when the robot moves, is not interfered by the noise of the working environment, and achieves a good measurement and calculation effect.

Description

technical field [0001] The invention relates to the technical fields of computer graphics and computer image processing, in particular to a structured light array designed for a robot, and a method for performing remote attitude calculation of a robot through the structured light array. Background technique [0002] Magnetic adsorption wall-climbing robots have important applications in the detection and maintenance of steel structures, and automated operations are their future development trend. Positioning and mapping the magnetic adsorption wall-climbing robot is a prerequisite for realizing automated operations. [0003] Space attitude is an important parameter for robot positioning. Currently, it is generally obtained through IMU (Inertial Measurement Unit, Inertial Measurement Unit) calculation. However, magnetic adsorption wall-climbing robots usually have adsorption structures such as magnetic wheels and magnetic arrays, which cause severe distortion of the surround...

AI Technical Summary

Problems solved by technology

[0005] However, when calibrating the wall-climbing robot, many types of calibration methods may be used, such as a variety of different spot colors or spot shapes, so it is necessary to set different LED lights or LED lights of different shapes, which improves the wall-climbing robot. The cost and energy consumption of the robot, and too many spots for calibration will also increase the computing power of spot recognition

Images