Research On Autonomous Navigation AGV Based On Lidar

With the advancement of science and technology,various types of robots are gradually playing an increasing role in various fields,taking on more and more tasks.More and more unmanned vehicles appear in the daily working environment.As a new direction,SLAM is also continuously improving with the development of unmanned aerial vehicles.As intelligent robots involve more and more market fields,there are higher requirements for the speed and accuracy of mapping.The two-dimensional lidar is used as the main observation sensor of the autonomous navigation vehicle,and the simultaneous laser positioning and mapping technology and path planning technology are analyzed,and a laser radar-based mapping and navigation scheme is designed using the robot operating system.First,in the robot operating system,establish a four mecanum wheel kinematics model,complete coordinate unification and grid map expression model,use the wheel encoder and IMU fusion method to replace the less accurate wheel encoder odometer,Established a positioning method based on odometer and lidar.Analyze the laser SLAM based on particle filter at this stage,and analyze and study the improved SLAM method based on Fast SLAM proposed in this paper.Combining the idea of genetic method,improve the resampling process,and through MATLAB simulation experiment,the improved method is compared with the traditional The Fast SLAM particle filter method is used to compare the error of pose and feature point.Through the comparison of experimental results,the improved algorithm is used to map the environment,and the map constructed is more accurate,the shape of the characteristic object is clearer,and the amount of calculation is smaller.Secondly,in the path planning research,in view of the shortcomings of ant colony algorithm that are prone to local optimization in the early stage,the advantages of the A*algorithm are used to improve the ant colony algorithm,and the evaluation function of the A*algorithm is introduced into the heuristic function to improve the ant colony path planning Algorithm,build a grid map simulation environment on the MATLAB platform for experiments.The experimental results show that the improvement improves the computational efficiency and path planning length of the ant colony algorithm.Finally,choose Raspberry Pi4 B,RPlidar A2 lidar,STM32 drive board,nine-axis IMU,and four mecanum wheels to build an autonomous navigation robot,and place SLAM mapping and path planning solutions on the autonomous navigation robot for actual environmental applications.Compare the SLAM method proposed in this paper with the path planning method and the comparison method,observe the pros and cons of the mapping effect and the decision path,and compare and analyze the obtained data.