Body robot dynamic obstacle avoidance pre-judgment system facing man-machine cooperation scene

CN121934572AActive Publication Date: 2026-04-28HANGZHOU ZHONGJIAN YUNTIAN TECH CO LTD
View PDF 4 Cites 2 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
HANGZHOU ZHONGJIAN YUNTIAN TECH CO LTD
Filing Date
2026-03-27
Publication Date
2026-04-28

AI Technical Summary

Technical Problem

Existing robot dynamic obstacle avoidance technology has difficulty accurately predicting human behavioral intentions in human-robot collaboration scenarios, resulting in a large deviation between the predicted trajectory and the actual situation. Furthermore, the fixed safety boundary strategy limits the robot's operational efficiency and the smoothness of collaboration.

Method used

A dynamic environment scene model is constructed using a perception fusion module. The trajectory prediction module predicts the future trajectory of the human body by performing pattern matching based on the historical trajectory database. The safety situation assessment module calculates the collision risk probability distribution map, dynamically adjusts the safety boundary, and generates collision-free candidate trajectories through a decision planning module. Real-time fine-tuning is achieved in conjunction with the motion control module.

Benefits of technology

It improves the accuracy and time foresight of human motion trajectory prediction, enables adaptive adjustment of robot safety strategies, reduces unnecessary motion interruptions, and enhances the smoothness and efficiency of human-robot collaborative operations.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN121934572A_ABST
    Figure CN121934572A_ABST
Patent Text Reader

Abstract

The invention, which relates to the technical field of robot intelligent obstacle avoidance, discloses a human-machine cooperation scene-oriented dynamic obstacle avoidance pre-judgment system for a body-equipped robot, comprising a sensing fusion module, a trajectory prediction module, a security situation assessment module and a decision planning module. The system constructs an environment model by fusing multi-source sensor data, and predicts a plurality of future potential motion trajectories of a human body based on historical trajectory database matching. The self-adaptive safety boundary is dynamically generated by calculating collision risk probability distribution of the robot and the dynamic obstacle. And finally, according to the risk map and the dynamic boundary, the task trajectory of the robot is re-planned, and a collision-free motion trajectory is generated. According to the technology, the accuracy of human body motion intention prediction is improved, and intelligent real-time adjustment of the safety boundary is realized, so that the safety and operation efficiency of man-machine cooperation are enhanced.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Obstacle avoidance method of robot

    CN121143438A

  • Dynamic obstacle pre-judgment obstacle avoidance system of service robot

    CN121254840A

  • Dynamic semantic SLAM (Simultaneous Localization and Mapping)-driven robot full-life-cycle navigation method

    CN121323632A

  • Train auxiliary driving control method and system based on four-eye vision

    CN121469517A

Cited By

  • Body robot obstacle avoidance review control method and device and body robot system

    CN122172704A

  • A collision-free trajectory dynamic programming method and system for a machining cell robot

    CN122253227A