Oiling robot visual guidance method based on images

Abstract

The invention discloses an oiling robot visual guidance method based on images. The method comprises the steps that based on the extracted automobile rounded rectangle fuel tank cap virtual corner features, through analysis of a pin-hole imaging model, a quantitative mapping relation model between tail end camera posture change and image space feature point change is built; the basic concept of animage Jacobian matrix and an image Jacobian specific expression form based on the model are disclosed, and a linear controller is arranged to conduct oiling task visual guidance simulated analysis based on the condition that camera calibration parameters exist; online estimation is conducted on an image Jacobian matrix in an uncalibrated visual guidance task through a kalman filtering algorithm;and an uncalibrated visual guidance oiling positioning system based on kalman filtering estimation is built. According to the method, simulated analysis is conducted on uncalibrated visual guidance oiling positioning under the MATLAB environment, and the effectiveness of the method for achieving the oiling task is verified. The simulation result positioning speed is 10 s or so, the speed of a robot does not exceed the safety speed, and the spatial path is approximately the best.

Description

technical field [0001] The invention belongs to the field of industrial robots and visual servoing, and in particular relates to an image-based visual guidance method for a refueling robot. Background technique [0002] With the rapid development of my country's economy and society, the continuous improvement of transportation infrastructure and the continuous improvement of residents' living standards, the number of motor vehicles in my country has increased rapidly. As a fuel supply place for automobiles, gas stations have gradually become an indispensable part of people's lives. By the end of 2018, the number of gas stations nationwide had exceeded 120,000. It can be seen that the number of gas stations in my country is huge and is still increasing, and the gas station market is huge. With the progress of society and the development of science and technology, robot technology has developed rapidly, robot automation equipment has gradually replaced traditional manpower t...

AI Technical Summary

Problems solved by technology

However, the target pose estimation is sensitive to camera and hand-eye calibration errors and the correctness of the target model, and the target may leave the camera's field of view during the terminal movement process, resulting in control failure. The 3D pose error estimation is time-consuming and has poor real-time performance.

Images