Research On Simultaneous Localization And Mapping Of Indoor Mobile Robot Based On Multi Sensor Fusion

The mobile robot is an important product of human stepping intelligent society.With the development of social productivity,mobile robot is widely used in various fields.On the one hand,It is necessary for mobile robot to realize effectively fusing multi-sensor,which is suitable for many complex scenes.On the other hand,the autonomous navigation is the foundation of finishing tasks autonomously.In this paper,the research and application of multi-sensor fusion SLAM technology and path planning algorithm are carried out based on the research background of mobile robot.Firstly,the motion model and sensor sensing model of mobile robot are studied.Secondly,the key technologies,such as point cloud matching,mapping and Cartographer algorithm framework,are studied and optimized by the research of multisensor fusion.As the motion model and the sensing model,firstly,the positioning principle of the differential motion of two-wheel model,the probability distribution of lidar and IMU preintegration model are introduced.Finally,the internal and external parameters of the sensor are calibrated by using the least square method,closed solution and discrete model respectively.the key technologies of robot SLAM are mainly divided into three parts: front-end point cloud matching technology,back-end loop optimization and mapping.The point cloud matching technology introduces ICP,NDT with PCL library and the manual ICP method.The error of point cloud experiment is divided.In the aspect of global path planning,both Dijkstra and traditional A* algorithm are introduced firstly,then the improved A* algorithm is proposed to compare the simulation results with other algorithm.In the aspect of local path planning,the DWA path planning is compared to verify the location impact of different parameters.In the aspect of improvement,we mainly improve and optimize the point cloud distortion from three aspects: the optimization end of Cartographer algorithm,distributed system design and enhanced data fusion.Finally,in order to verify the effectiveness of the improved Cartographer algorithm,the simulation experiment based on the gazebo platform is carried out,and then the composition effect is compared in the physical test,and the Gmapping algorithm is added into comparison.Meanwhile,in order to verify the effectiveness of the fusion multi-sensor,the design experiment can add different sensor information to the simultaneous in the same environment,and compare the building effect.Not only reduces the interference of the independent variables,but also enhances the comparability of the algorithm.