Robot curved surface motion positioning method and system

Abstract

The invention discloses a robot curved surface motion positioning method and system. According to the method, an RGB-D depth camera is adopted for acquiring a color image and a depth image of a curvedsurface of the position in front of a robot during movement, a triaxial angular velocity value and a triaxial acceleration value of an inertial measurement unit are recorded, a rotation matrix and atranslation vector obtained through feature matching are optimized, and point cloud registration is conducted according to the depth image; and a registered point cloud is divided into approximately planar grids, the robot motion velocity is calculated according to an output value of a wheel encoder, and the robot displacement in the grid is calculated through a plane odometer method. Through theadopted depth point cloud registration method, all robot curved surface motion scene information is fully utilized, the real-time speed of the robot is obtained through the encoder, the position positioning precision and the posture positioning precision of the robot during movement on the curved surface for a long time are improved, and the anti-interference ability is high.

Description

technical field [0001] The invention relates to a computer vision technology, in particular to a robot curved surface motion positioning method and a motion positioning system thereof. Background technique [0002] With the development of robot technology and the increase of labor costs, robots can replace manual cleaning of high-rise building exterior walls, surface inspection and maintenance of large structural parts, welding and inspection of large equipment such as large ships, which can improve production efficiency, reduce manufacturing costs, and reduce labor costs. Labor intensity and operational risk. The industrial wall-climbing robot consists of a working platform and a mobile chassis that can absorb wall movement, and can move and work in dangerous environments such as vertical walls. The industrial wall-climbing robot needs to adjust the angle and position of the working unit according to its own posture and position information, such as welding, grinding, surf...

AI Technical Summary

Problems solved by technology

Chinese patent document CN201110417967 discloses a mobile robot motion attitude estimation method based on machine vision and inertial navigation fusion; A strapdown inertial navigation system / visual odometer combined navigation method uses the fusion of binocular cameras and inertial navigation, and binocular cameras are not effective in matching the surface features of smooth structural parts; Chinese patent document CN201710320077 discloses a method based on The robot positioning and map construction system based on monocular vision and IMU information uses the triangulation method to extract the depth of feature points. When the number of key frame feature points is small, the positioning error is relatively large; Chinese patent document CN201711050182 discloses a mobile robot based on monocular The terminal AR method uses a combination of a monocular camera and an inertial measurement unit, in which the visual motion parameters are obtained by the optical flow method, which is harsh on the lighting conditions and has a large repeat positioning error; Chinese patent document CN201711077826 discloses a visual inertial navigation information based The fusion positioning method and system adopts the combination method of RGB-D camera and inertial measurement unit, only uses the depth value of the feature point, and the effect is general in the smooth surface position positioning

[0005] The technology disclosed above is not suitable for long-term position positioning and attitude positioning of industrial robots on the surface of large structural parts or large equipment

Images