Research On Docking Path Planning Of Unmanned Boat Based On Multi-source Fusion Positioning

Unmanned Surface Vessel(USV)is widely used in military reconnaissance,environmental monitoring,scientific research experiments,resource exploration,patrol search and rescue and other tasks.In order to improve the efficiency of unmanned rescue vessels’ retractio,and thus the efficiency of rescue missions,this paper proposes an recovery dock device to recover the USV.There are difficulties in positioning and path planning in the docking process.The accuracy of USV positioning system is low,hence it cannot meet the requirements of automatic docking.In addition,the current path planning algorithm for USV cannot meet the complicated automatic docking task.In order to improve the positioning accuracy,this paper establishes a tightly coupled fusion positioning model based on the studies of USV positioning technology.Then this paper proposes a global navigation satellite system(GNSS)and Ultra Wide Band(UWB)cooperative positioning method based on the Extended Kalman Filter(EKF).Simulation and actual test results show the error of GNSS/UWB positioning system is reduced by about 30% compared with the traditional UWB positioning method.In order to solve the problem of automatic docking path planning,this paper models the docking path planning as a multi-constraint and multi-optimization problem,and proposes an algorithm that combines global path planning and trajectory generation to decouple multiple constraints and multiple optimization goals.Based on the improved A* and RDP algorithm,the global waypoints with path length optimization are planned to solve the collision and yaw limitation.The global waypoints is used as the constraint condition of the trajectory generation model,and the minimum snap bezier curve is generated by the model,which satisfies the constraints of continuity and kinematic parameters.The simulation experiment results show that compared with the traditional A* algorithm,the improved A* and RDP algorithm proposed in this paper can shorten the path length by about 4%-9% on average.The trajectory generation algorithm can generate a smooth trajectory that meets the constraints of continuity and kinematics parameters for USV.By designing system hardware and software,the USV test platform is built,and the actual tests are carried out over the lake and sea.Test results show that the GNSS/UWB fusion positioning system and path planning algorithm of this paper can meet the requirements of USV docking task.