| In recent years,more and more enterprises have widely applied AGV as the key automation equipment of modern logistics in indoor logistics system for the goal of improving logistics efficiency and reducing labor cost.The navigation function is the core ability of AGV to complete the handling task,but in the complex environment of indoor logistics,the traditional single navigation methods are obvious imperfect.Therefore,a suitable,efficient and safe and reliable navigation method needs to be studied.In this thesis,by analyzing the advantages and disadvantages of various navigation methods,combining free path and fixed path navigation methods,an indoor AGV navigation method based on virtual road signs and lidar is proposed,and the establishment and identification of virtual road signs,based on virtual road signs the guidance and virtual road sign navigation method fused with A* algorithm are researched,and the autonomous navigation in the logistics dynamic environment is realized.The main work is as follows:(1)Multi-layer cost maps are studied and virtual road signs are designed to meet the needs of navigation,and the creation of virtual road signs is realized.By analyzing the principle of the multi-layer cost map that AGV navigation relies on,and creating a custom virtual road sign map layer;then according to the characteristics of 2D lidar mapping and combined with the needs of navigation,the virtual road sign is designed and completed in the custom layer.Creation of virtual road signs.(2)A virtual road sign recognition algorithm and a virtual road sign guiding AGV method are proposed,so that the AGV can recognize the virtual road sign and guide the AGV movement according to the road sign information.By obtaining the cost value in the local cost map in real time,and through the contour detection algorithm based on the sliding point,the contour features are compared with the designed road signs,so as to realize the identification of virtual road signs.Then analyze the kinematics model and speed control instructions of AGV,and combine the motion speed control instructions with virtual road signs to realize the motion guidance of AGV.(3)A global path planning and navigation method based on virtual road signs is proposed to realize the autonomous navigation of AGV in the logistics dynamic environment.The entire navigation process of AGV is decomposed into free path navigation + virtual road sign navigation + free path navigation,and free path navigation is used for the stage from the starting point and the end point to the logistics channel,so as to realize the flexible arrangement of the starting point and the end point;in the dynamic area of logistics,the road signs guide Introducing the AGV to standardize the movement in the logistics channel can improve the safety and reliability of the AGV movement process and maintain a high speed.Two local navigation modes,motion correction and avoidance of dynamic obstacles,are also proposed to improve the stability and reliability of AGV navigation.(4)The above research and methods are verified by experiments on the recognition algorithm of virtual road signs,virtual road signs to guide AGV motion and the navigation method based on virtual road signs.The experimental results show that: the average recognition time of virtual road signs is 93 ms,which meets the real-time requirements of navigation;the position and angle deviations of AGV motion guided by virtual road signs are 2.08 cm and 1.13°,respectively,which can meet the navigation needs of AGV;The navigation method of road signs is able to carry out effective navigation,and compared with the traditional A* algorithm,the navigation efficiency,safety and reliability are improved by6.4%. |