Research On Autonomous Navigation Of Unmanned Surface Vessel Based On Multi-sensor Fusion

With the advent of the era of artificial intelligence,intelligent robots have been rapidly developed.As a branch of intelligent robots,unmanned surface vessel have been widely used in national defense and civil fields.In this context,aiming at the current situation of autonomous navigation of unmanned surface vessel,this topic designs and implements an autonomous navigation system of unmanned surface vessel based on multi-sensor fusion.The system has the function of planning the path independently,and uses the data of encoder,IMU and LIDAR for fusion positioning and mapping,so as to avoid the limitation of a single sensor and improve the environmental adaptability of the unmanned surface vessel during navigation.Firstly,based on the research status of multi-sensor unmanned surface vessel at home and abroad,the autonomous obstacle avoidance technology,SLAM technology and path planning technology of unmanned surface vessel were introduced,and the main research content of this topic was proposed.According to the operating environment and functional requirements of the unmanned surface vessel,the overall scheme was designed,and the unmanned surface vessel system was divided into propulsion system,hardware system and software system.According to the experimental requirements,the three-dimensional model of the body structure of the unmanned surface vessel was designed by Solid Works software.According to the speed requirements of the unmanned surface vessel,the power propulsion system adopts the scheme of DC brush motor with forward and reverse OARS.In order to ensure the motion control of the unmanned surface vessel,the differential propulsion kinematics model of the unmanned surface vessel was calculated.The hardware system of the unmanned surface vessel was built and its hardware equipment was introduced.The TCP/IP data communication system based on WIFI network was built,and the communication system was tested by picture transmission.The architecture of the software system was designed,and the robot operating system is introduced.Secondly,the positioning and mapping scheme of the multi-sensor fusion of the unmanned surface vessel was studied.In this thesis,the adaptive Monte Carlo algorithm was used to fuse the data of LIDAR,encoder and IMU to locate the unmanned surface vessel.Firstly,the data of the photoelectric encoder was analyzed,and the dead reckoning model of the unmanned surface vessel was deduced in order to obtain the pose at a certain moment.This thesis introduces the ranging principle of LIDAR,deduces the pose transformation of point cloud data,and establishes the mathematical model from initial data to positioning parameters.Then the unmanned surface vessel positioning scheme design,scheme extended Kalman filtering algorithm was adopted in a fusion of encoder,innovative marketing position data and plan 2 will use the adaptive monte Carlo algorithm was used to merge encoder,IMU,laser radar data,positioning algorithm fusion principle of two kinds of schemes was introduced and calculated.After completing the positioning of the unmanned surface vessel,it was necessary to map the environment.In this thesis,Cartographer algorithm based on graph optimization was used to map the environment.On the basis of completing its own positioning and raster map,the effect of global path planning algorithm Dijkstra and A* algorithm was compared.Hybrid path planning algorithms,including A* algorithm and DWA algorithm,were used and simulated respectively.Finally,the autonomous obstacle avoidance system of unmanned surface vessel was tested experimentally.In the positioning accuracy experiment,by comparing the fusion scheme based on encoder and IMU and the fusion scheme based on LIDAR,encoder and IMU,the results show that the positioning data based on multi-sensor fusion was closer to the ideal path of the unmanned surface vessel.The effect of Cartographer algorithm based on graph optimization was verified,and the raster map built by it was close to the contour of the campus lake.Experimental results show that the trajectory of the unmanned surface vessel was close to the ideal trajectory line.Under the influence of wind and waves and other environmental factors,it can become stable,and it can timely make obstacle avoidance actions when encountering obstacles,so as to realize the stable and reliable operation of the unmanned surface vessel.