基于多传感器信息融合的AGV自主无轨导航技术

Autonomous Orbitless Navigation Technology for AGV Based on Multi-sensor Information Fusion

传统的物料搬运AGV机器人不能进行自主规划行走路径,也不能满足智能制造和数字化工厂对AGV进行科学调度、自动避障、安全可靠且不发生阻设的要求。因此,文章基于多模态传感器数据处理、SLAM地图构建等多重信号融合的原理和方法,提出了一种多传感器信息融合的AGV自主无轨导航技术,设计了AGV实物样机,并进行实验验证。结果表明,在无轨复杂场景情况下,AGV机器人的导航定位精度优于1 mm,重复定位精度优于2 mm,满足智能制造和数字化工厂对AGV的自主导航控制要求。

Traditional material handling AGV robots cannot independently plan their walking paths,nor can they meet the requirements of intelligent manufacturing and digital factories for scientific scheduling of AGV,no obstruction,automatic obstacle avoidance,and safety and reliability.Based on the principles and methods of multi-signal fusion such as multi-modal sensor data processing and SLAM map construction,the paper proposes an autonomous trackless navigation technology for AGV based on multi-sensor information fusion and designs a physical prototype of AGVs,and the experimental verification has been conducted.The results show that the navigation and localization accuracy of the AGV robot is less than 1 mm and the repeat localization accuracy is less than 2 mm in the case of trackless complex scene,which meets the the requirements of intelligent manufacturing and digital factories for autonomous navigation control of AGV.