Image based simplified robot visual servo control method

Abstract

The invention discloses an image based simplified robot visual servo control method, which is applied to the field of robot intelligent control, and solves the problems that an existing visual servo method is great in calculation quantity and low in control precision. The method comprises the following steps: obtaining an expected image, and processing the expected image to obtain image characteristics of a workpiece as expected characteristics; testing a mapping proportion between an image space and a robot Cartesian space through multi-time exploratory movement; and acquiring and processing the image information under a current position when the robot is on any position capable of acquiring the image information of the workpiece, and controlling the robot to perform next-step motion according to a deviation value by calculating a deviation value between the image and the expected image according to the mapping proportion. The method disclosed by the invention not only reduces calculation quantity in a control process, and also improves real-time property of the system.

Description

Technical field [0001] The present invention belongs to the field of robot intelligence control, and in particular, the present invention relates to a flat visual robotic control technology. Background technique [0002] At present, the underlying industrial robot works in the industrial site, generally controlling the robot by teaching all the actions of the robot, but such control is strictly restricted by the robot's ability to grab the target object, meet the current industry Moisturized, intelligent work demand. With the advancement of image processing and robot related technologies, machine vision technologies are increasingly used to guide industrial robots to complete the flexible work tasks, greatly increase the autonomous and flexibility of industrial robots. Visual servo is mainly divided into two forms, location and image based on images. [0003] Location-based visual servo requires a large number of debugging and calibration work in the previous period, but many ind...

AI Technical Summary

Problems solved by technology

However, this method has the following problems: (1) the image acquisition equipment used in this method is a medical endoscope, and the method is mainly used in the medical field rather than the industrial site; (2) the tracking point and the target point of the method are only Position parameters have no attitude parameters. In the industrial field, when the robot grabs the workpiece, the attitude needs to be considered; (3) Since this method is used in the medical field, it has high requirements for the flexibility of the control process, and needs to be real-time through feedback. Evaluate the tracking performance and update the rate factor, which is computationally intensive

Images