Development Of AGV Indoor Navigation Software Based On Laser Radar

With the rapid development of intelligent manufacturing,the demand for high-end intelligent equipment is growing.As an automatic and intelligent equipment,AGV plays an important role in many industries.The navigation technology is the important technology of AGV.AGV accurately senses environmental information,and building environmental maps is the basis for autonomous navigation.Therefore,this paper uses laser radar as a sensor to research and develop the AGV indoor navigation software based on graph-based SLAM and path planning based on grid map.This paper firstly designs a three-layer software architecture from hardware driver layer to application layer,and then analyzes and establishes the motion model,odometry model and lidar-based measurement model of differential AGV.In this paper,the graph-based SLAM is deeply analyzed.The system block diagram of the indoor mapping module is designed.The scan matching is implemented based on CSM and optimization method.The loop closure detection is realized based on multi-resolution pre-processing grid and branch-and-bound approach.The nonlinear least squares method is used to implement pose optimization.Based on the above sub-modules,the indoor mapping function is realized.And a visual interface for the mapping process is developed.In this paper,the AGV walking cost and hierarchical map based on grid map are deeply analyzed.The system block diagram of indoor navigation module is designed.The global path planning is implemented based on A* algorithm,the local path planning is implemented based on dynamic window method;based on the above sub-modules,the indoor mapping function is realized.And a visual interface for the navigation process is developed.In this paper,the AGV indoor navigation software is developed under the Ubantu16.04 system,and the navigation software is tested on the differential AGV.The test results show that a globally consistent grid map can be obtained,the AGV can avoid obstacles and safely reach the target point,achieving autonomous navigation.