Research On Multi-target Autonomous Navigation Technology Of Intelligent Logistics Vehicle Based On ROS

With the rapid development of science and technology,intelligent logistics vehicle has become an important element in intelligent logistics,playing a vital role in the handing process of manufacturing,express delivery,e-commerce and other industries.Autonomous navigation is the top priority of intelligent logistics vehicles,and path planning is an important branch of autonomous navigation.At present,there are many kinds of intelligent logistics vehicles with various uses,but the fact that each of them has its own platform makes development more difficult.ROS is an open source software platform with strong versatility,good compatibility and support for multi-language programming.In addition,the current path planning is mostly focused on a single goal point,which cannot meet the current multiple tasking needs of intelligent logistics.Therefore,the multi-target navigation technology of intelligent logistics vehicles based on ROS platform are studied in this dissertation.Firstly,the basic principle and classification of SLAM algorithm are studied,and the system model of SLAM is constructed.On this basis,the basic principle and process of RBPF-SLAM algorithm are further studied.The map of simulation environment is constructed by using this algorithm,and the prior map is successfully constructed,which verifies the feasibility of RBPF-SLAM algorithm.Secondly,the algorithm of path planning is mainly studied.Global path planning and local path planning are studied respectively.Aiming at the global path planning algorithm,the traditional A* algorithm is mainly studied.In order to solve the problem of too many searching nodes and broken lines,three improvements are made,search directions are reduced,redundant nodes are deleted and the evaluation function is improved.The simulation result shows that the improved A* algorithm greatly reduces the number of traversal nodes,decreases the inflection point of the path,shorten the length of the search path and makes the path smoother,which greatly improves the efficiency of global path planning.Aiming at local path panning,the basic characteristic and principle of ant colony algorithm are studied.MATLAB is used for simulation experiments,and combined with the improved A* algorithm to complete the path planning between two adjacent points,the TSP problem of 10 target points is solved,and the optimal traversal sequence and shortest path of multi-target points are obtained,which laid a foundation for the realization of ROS multi-target points navigation in the future.Finally,the principle of laser rangefinder which required for navigation in the hardware part and three levels and characteristics of ROS basic architecture in the software are studied.The Ubuntu and ROS system are installed,and the simulation experiment of multi-target navigation of intelligent logistics vehicle is realized.The simulation environment is established in Gazebo software,the simulation map is constructed with RBPF-SLAM,and the path planning is carried out with the improved A* algorithm and ant colony algorithm.The multi-target navigation of the intelligent logistics vehicle is completed on the ROS platform,which proves the feasibility of the navigation technology studied in this dissertation.