A method for correcting the spatial positioning accuracy of prosthetic objects

Abstract

The invention discloses a method of correcting the positional accuracy of a target space for an artificial limb. The method comprises the steps of initially locating a series of targets by using a method of realizing carrying out space positioning to a random target by virtue of the artificial limb with multi-degree of freedom so as to obtain an integration of a series of target point positions, and leaving the integrations a training input sample set of an artificial neural network; precisely measuring an integration of the practical three-dimensional coordinate values of a series of targets, leaving the integration of the practical three-dimensional coordinate values as the training input sample set of the artificial neural network, building the artificial neural network, and enabling an output value of the artificial neural network to be close to a real value close to a target after the fully training. The actual output of the artificial neural network is close to the true value when the training input sample integration and parameters close to the training input sample integration are input into a network, so that a precision correcting model for carrying out space positioning on a random target by an upper limb artificial limb with multi-degree of freedom or a robot can be formed to effectively correct the positioning precision of the target by the artificial neural network.

Description

technical field [0001] The invention relates to an upper limb prosthesis for a handicapped with broken shoulder, in particular to a precision correction method for the spatial positioning of a random target of the prosthesis. Background technique [0002] In the applicants invention patent "A method for realizing the spatial positioning of random targets by multi-free prosthesis" (ZL201010280508.3), a method for the upper limbs of the "shoulder dislocation" type of disabled people in an unstructured environment is disclosed. Prosthesis, especially a multi-degree-of-freedom upper limb prosthesis that can realize compensatory functions, is a method for spatially positioning random manipulation targets. During the use of this method, it is found that when a target is initially positioned, its three-dimensional coordinate value p=[p x p y p z ] T , because of the difference in the size of the wearer, the error in the measurement of the body structure, the error in the inst...

AI Technical Summary

Problems solved by technology

During the use of this method, it is found that when a target is initially positioned, its three-dimensional coordinate value p=[p x p y p z ] T , because of the difference in the size of the wearer, the error in the measurement of the body structure, the error in the installation position of the sensor, etc., the error between the coordinate value of p and the real value is often relatively large, which affects the description accuracy of the relationship between the relevant coordinate systems , the error accumulation described by the relationship between multiple coordinates will eventually amplify and affect the target positioning accuracy, directly using its value will affect the operation effect of the upper limb prosthesis, which is not conducive to the accurate operation of the upper limb prosthesis on the target

Images