Lost Water Target Search Technology Based On Waterjet Propelled Unmanned Surface Vehicles

In the context of China’s strategy of building an ocean power and the Silk Road Economic Belt and the 21st-Century Maritime Silk Road,with the increase in maritime economic and trade activities,the occurrence of accidents and emergencies is significantly increased.Existing manned searching forces are faced with challenges of large-scale and long-term search operations in response to emergency search and rescue(SAR)missions at sea.To solve the above-mentioned problems,this thesis designs a waterjet propelled unmanned surface vehicle(USV).Based on stochastic particle simulation,a high-probability search area is defined to increase the probability of containment(POC)of a person who falls into the water in a specific area.Then,take the probability of detection(POD)and energy consumption as optimization targets,the sweep direction and sweep spacing are optimized by the parallel selection genetic algorithm.Considering the maneuverability constraints of the USV,the optimized searching path is proposed.Combined USV hardware platform with ROS and Gazebo simulation solutions,the applicability and traceability test of the USV is carried out.In this thesis,the probability of success(POS)is increased by optimizing POC and POD.The contents of the thesis are listed as follows.Firstly,a USV with an inertial navigation unit and the power unit is designed and built for the surface water searching missions.The USV thrust versus throttle output curves and energy consumption models were determined through field tests.Secondly,the drift model of the distressed target based on the leeway model is determined by the initial position distribution of the distressed target and the wind and flow field disturbances.Based on the Monte Carlo method,the search area is determined by stochastic particle simulation.The drift model in this thesis was verified by using the drift test data of distressed targets at sea from papers.The searching area determined in this thesis was decreased under POC of 100%,which is less than that from the analytical method.Thirdly,compared with the existing water search methods,according to the motion characteristics and energy consumption characteristics of the USV,the parallel line scan search method was determined to cover the search area.In the area coverage path planning,there are multi-objective optimization functions of POD and energy consumption.To achieve the goal of the minimum energy consumption and the maximum POD of the USV in the search area the parallel line search direction and search spacing are optimized by the parallel direction genetic algorithm.The turning path is sorted by the relationship between the sweep spacing and the turning circle diameter of the USV.Then the performance of the USV in the heading angle tracking and path tracking is improved.Thus,an optimized path for parallel line scanning is proposed.Finally,the ROS and Gazebo simulation environment that can load the wind and flow data are built to simulate the area coverage path planning of the USV in the search area.The simulation results show that the parallel line scan optimization path can improve the coverage,working path length,path efficiency,and POD in the search area.Besides,it reduces the heading angle tracking delay time,heading angle in the non-working path,total heading angle variation,path tracking error,and energy consumption.This thesis studies the methods of improving the POS from search vehicle,search area,and path planning aspects in surface searching technology and gains some achievements.A USV platform for the surface searching mission was built.The search area of the lost water target was determined by computer simulation methods,and the surface searching path planning was optimized by a genetic algorithm that improved the POD in the search area.Through that,the POS was increased.