Multi-degree-of-freedom bionic manipulator control method

By establishing the kinematic model and dynamic relationship of the bionic manipulator, and combining inertial measurement, computer vision and surface electromyography sensing modes, and using deep neural networks for multi-source information fusion, the shortcomings of existing manipulators in dexterity and multi-degree-of-freedom control are solved, and high-precision bionic manipulator control is achieved.

CN122442733APending Publication Date: 2026-07-24BEIHANG UNIV
0 Cites 0 Cited by

Patent Information

Application Number
CN202610880200.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-06-17
Publication Date
2026-07-24

AI Technical Summary

Technical Problem

Existing robotic arms lag significantly behind human hands in terms of dexterity, compliance, and multi-degree-of-freedom control, making it difficult to meet the needs of complex operations and human-machine collaboration.

Method used

The kinematic equations of a multi-degree-of-freedom bionic manipulator are established, a pre-trained pattern recognition model is constructed, and synchronous perception is achieved using three heterogeneous sensing modalities: inertial measurement, computer vision, and surface electromyography. Multi-source information fusion regression is realized based on deep neural networks to output target control commands for each joint of the manipulator, and high-precision control is achieved through real-time closed-loop control.

Benefits of technology

It achieves high-precision control of multi-degree-of-freedom bionic robotic hands, and is suitable for scenarios such as intelligent prostheses, remote teleoperation, and dexterous operation of humanoid robots.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of robots, and provides a multi-degree-of-freedom bionic manipulator control method. According to the method, firstly, a kinematic model and a dynamic relation of the bionic manipulator are established, and on the basis, the multi-modal training and real-time control method for the 20-degree-of-freedom bionic manipulator is provided. According to the method, three heterogeneous sensing modes of inertia measurement, computer vision and surface myoelectricity are comprehensively utilized, the hand motion intention of a human operator is sensed synchronously, fusion regression of multi-source information is achieved based on a deep neural network, and target control instructions of all joints of the manipulator are output. In the training stage, an inertial measurement system is used for providing a high-precision joint angle true value; and in the deployment stage, vision and surface myoelectricity are mainly adopted as network input, and real-time closed-loop control is realized in combination with manipulator end feedback, so that high-precision control of the multi-degree-of-freedom bionic manipulator is realized.
Need to check novelty before this filing date? Find Prior Art