双AGV避障系统设计与分析

Design and analysis of dual AGV obstacle avoidance system

通常自动导引车(AGV,automated guided vehicle)根据其分配的任务在仓库中协同执行多项工作,而一些障碍物,如墙壁、杆子、货架和其他物体会干扰它的移动。因此,AGV不仅需要具有避开障碍物的能力,还需要收集提升其整体性能的位置信息。本研究使用机器人操作系统(ROS,robot operating system)平台在两台AGV之间建立了无线通信,利用激光雷达扫描数据并采取相关的避障方法,开发了AGV协作系统。通过进行3种不同场景的AGV运动实验,验证了该协作系统在两辆车交会时能够共同完成任务,应用统计中的方差分析方法,确定了影响自动导引车最终位置误差的主要因素。从单因素方差分析的结果来看,不同场景可能会对x坐标和y坐标的误差产生单独的影响。

AGVs collaborate to perform multiple tasks in the warehouse based on their assigned tasks,but some obstacles,such as walls,poles,shelves and other objects can interfere with the movement of AGVs.Therefore,AGVs not only need to have the ability to avoid obstacles,but also need to collect location information to improve their overall performance.This study established wireless communication between two AGVs using the ROS platform,and developed an AGV collaboration system using LiDAR scanning data and relevant obstacle avoidance methods.By conducting AGV motion experiments in three different scenarios,it was verified that the collaborative system can jointly complete tasks when two vehicles intersect,and applied statistical analysis of variance to determine the main factors affecting the final position error of the AGVs.From the results of one-way ANOVA,it can be seen that different scenarios may have a separate impact on the errors in the x and y coordinates.