Uncalibrated hand-eye coordination fussy control method based on support vector regression (SVR) learning

Abstract

The invention relates to an uncalibrated hand-eye coordination fussy control method based on support vector regression (SVR) learning. The method comprises the steps of recording the moving conditions of a paw of a robot and an object by a camera, projecting the moving conditions to an image plane, marking the projection positions of the paw and the object in the image plane by adopting an image processing and object identification method, inputting the position difference between the paw and the object to an SVR-based fussy controller designed in the invention as an input quantity, and outputting the change quantity of an arm joint of the robot by the controller so as to control the paw to move to the tracked object. By adopting the method, the quantity of samples is greatly reduced, a few steps are adopted and better effects are achieved in the control, meanwhile, the method also has the characteristics of good self-learning and generalizing capability and language interpretability. The design of the fussy controller based on a fuzzy basic function not only provides convenience for analyzing the similarity of functions, but also provides an effective path for combining expert knowledge and fussy rules generated by data information.

Description

technical field [0001] The invention relates to a robot hand-eye coordination control method, in particular to a calibration-free hand-eye coordination fuzzy control method based on support vector regression machine learning. Background technique [0002] Support Vector Regression (SVR) is a new machine learning algorithm developed and proposed by Vapnik et al. on the basis of statistical theory in the mid-1990s. As the core content of statistical learning theory, SVR can better solve the small sample learning problem, and support vector regression machine has become a research hotspot in machine learning modeling and optimization. [0003] Since the robot itself is a nonlinear and strongly coupled system, there are often various unknown parameters, and there is a complex nonlinear relationship between the end-effector image information acquired by the camera and the angles of each joint. Not readily available, and where such models exist, they often correspond to complex c...

AI Technical Summary

Problems solved by technology

For the visual mapping model in robot visual servoing, if the traditional intelligent algorithm is used for modeling, a large number of samples in the robot motion space need to be trained offline, but the number of samples is often limited, especially for complex motions with high degrees of freedom. As the number of dimensions increases, the training samples will grow geometrically, and the samples that can be obtained in practice become very limited. Therefore, it is difficult to guarantee the infinite number of samples in practice. The performance of a theoretically excellent learning method in practice may be as low as unsatisfactory

Images