Mechanical arm grabbing intelligent control method and system based on online pose correction

By equipping the end effector of the robotic arm with a mechanical sensor and a binocular vision sensor, and combining them with machine learning algorithms, the problem of inaccurate error correction of the robotic arm was solved, and accurate error correction and compensation were achieved in complex environments.

CN122008244BActive Publication Date: 2026-06-09GUANGDONG SHUNLI TECH CO LTD
View PDF 2 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
GUANGDONG SHUNLI TECH CO LTD
Filing Date
2026-04-07
Publication Date
2026-06-09

AI Technical Summary

Technical Problem

In existing technologies, error correction of robotic arms relies on visual servoing or visual guidance, which suffers from lighting and occlusion issues, resulting in inaccurate error monitoring, inability to adapt to complex industrial environments, and poor correction effects.

Method used

By detecting the mechanical feature matrix through a mechanical sensor configured at the end of the robotic arm, and performing registration analysis in conjunction with a binocular vision sensor, a drift error predictor is constructed to predict and correct errors. The model is then optimized using machine learning algorithms to achieve adaptive adjustment of errors.

Benefits of technology

It improves the stability and accuracy of error prediction during the robotic arm's grasping process, enhances the system's adaptability in complex environments, and achieves precise error correction and compensation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122008244B_ABST
    Figure CN122008244B_ABST
Patent Text Reader

Abstract

This invention discloses an intelligent control method and system for robotic arm grasping based on online pose correction, relating to the field of intelligent robotic arm control. The method includes: detecting a mechanical feature matrix using a mechanical sensor at the end of the robotic arm to predict drift errors; acquiring binocular images using a binocular vision sensor on the robotic arm, performing registration analysis to obtain registration drift error and registration drift reliability; verifying the obtained drift consistency parameters, combining them with the registration drift reliability, and labeling them to obtain a drift prediction training data set; analyzing the consistency of the label information to obtain label stability parameters, fusing the predicted drift error and registration drift error to obtain a processed drift prediction training data set, and using this data for robotic arm correction control and updating the drift error predictor. This invention solves the problem in existing technologies where inaccurate error monitoring of detection equipment leads to excessive errors during correction, poor correction effects, and inability to adapt to complex industrial environments.
Need to check novelty before this filing date? Find Prior Art