Research On Unmanned Distribution Vehicle Navigation Algorithm Based On ROS

With the rapid development of modern intelligent logistics and transportation technology,unmanned distribution vehicles developed based on Robot Operating System(ROS)have become an effective solution for rapid delivery of goods in the logistics and transportation industry.The unmanned delivery vehicle is a product of the combination of the new retail field and the Internet of Things technology.It has the functions of environmental recognition and positioning,optimal path planning,and real-time obstacle avoidance.It is the current research hot-spot in the field of autonomous mobile robots.This article focuses on the improvement and optimization of the navigation algorithm for unmanned delivery vehicles to achieve a safer,more accurate and more efficient delivery of unmanned delivery vehicles.In this paper,the ROS robot software framework is used to embed the navigation algorithm code designed and implemented by independent research into the unmanned delivery vehicle equipped with ROS,combined with the lidar sensor used to obtain external environmental information and the simultaneous positioning and map generation for environmental recognition Algorithm,design and implement autonomous navigation algorithm for unmanned delivery vehicles.Autonomous navigation function is the core function in the process of delivery of goods by unmanned delivery vehicles.First of all,this article specifically elaborates the research background and practical significance of autonomous navigation of unmanned distribution vehicles,analyzes and summarizes the research status at home and abroad,and proposes the overall design and focus of the autonomous navigation algorithm system of unmanned distribution vehicles based on ROS Research on global path planning algorithms and local path planning algorithms.Aiming at the accuracy of environment recognition and positioning and the optimal of path planning during autonomous navigation of unmanned delivery vehicles,improved and optimized simultaneous positioning and map generation algorithms,global optimal path planning and local real-time avoidance for specific environments The barrier algorithm enables unmanned delivery vehicles to deliver items safely and efficiently.Aiming at the environmental recognition and positioning problems in the delivery process of unmanned delivery vehicles,this paper improves the Gmapping simultaneous positioning and map construction algorithm so that the improved algorithm can better complete the positioning and map construction functions in a specific environment.The test results show that,by embedding improved and optimized simultaneous positioning and map construction algorithms,the unmanned delivery vehicle can make full use of the environmental information collected by the lidar sensor,quickly and efficiently construct the unmanned delivery vehicle distribution scene map,and can accurately and real-time Positioning provides the possibility for the optimal path planning of the unmanned delivery vehicles.At the same time,focusing on the characteristics of low speed and safety in the process of unmanned delivery vehicles,it focuses on improving the existing A* algorithm to achieve the global optimal path planning function of unmanned delivery vehicles,and focuses on improving the optimized dynamic window algorithm(DWA)Algorithm to realize the local real-time obstacle avoidance function of unmanned delivery vehicles.Finally,the improved and optimized simultaneous positioning and map construction algorithm,the key improved A* algorithm and DWA algorithm are integrated into the autonomous navigation system of the unmanned delivery vehicle,and tested by MATLAB,GAZEBO and other simulation software.The test results show that: no one After the distribution vehicle integrates the improved and optimized autonomous navigation algorithm,it has the ability to safely,accurately and efficiently implement the autonomous navigation function in different distribution scenarios.