A multi-agv cooperative perception method

By integrating information from infrastructure sensors and sensors on the AGV, and employing techniques such as multi-sensor target fusion and Kalman filtering, a global dynamic scene map is generated. This solves the problems of small perception range and occlusion in AGVs, and improves the reliability and safety of path planning.

CN115755887BActive Publication Date: 2026-06-23XI AN JIAOTONG UNIV
View PDF -1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
XI AN JIAOTONG UNIV
Filing Date
2022-10-26
Publication Date
2026-06-23

AI Technical Summary

Technical Problem

Existing AGV sensing systems have a small field of view, making it difficult to perceive the full picture of obstacles and easily obstructed by objects such as shelves. Furthermore, existing collaborative sensing systems fail to effectively share environmental information, resulting in unreliable and unsafe path planning.

Method used

By integrating the perception information from infrastructure sensors and sensors on AGVs, target tracking is performed using multi-sensor target fusion, Kalman filtering, and data association. Real-time scene information updates are combined with dynamic obstacles to generate a global dynamic scene map.

Benefits of technology

It expands the perception range of AGVs, improves the reliability and safety of path planning, reduces the probability of false detection of obstacles, and realizes efficient navigation in complex environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115755887B_ABST
    Figure CN115755887B_ABST
Patent Text Reader

Abstract

A multi-AGV collaborative perception method, comprising the following steps: S100: perception and measurement of infrastructure sensors and AGV-mounted sensors; S200: multi-sensor target fusion, fusing 3D target detection results from different single AGVs and infrastructure sensors to obtain target fusion results; S300: target tracking after multi-sensor target fusion, mainly including kalman filtering and data association; S400: combining dynamic obstacles to update real-time scene information dynamic obstacles and occupancy grid information. The method solves the problems of small AGV field of view, difficulty in perceiving the overall picture of obstacles, and easy to be blocked by objects such as shelves.
Need to check novelty before this filing date? Find Prior Art