Research On Autonomous Task Planning And Assignment Method Of Multiple UUVs For Searching Underwater Targets

With the development of unmanned technology and the exploitation of oceanic resources,UUV has been widely used in the field of ocean observation,detection underwater,area guard,etc.Especially in recent years,UUV has been used for many times to search for the wreckage of seabed wrecked targets because of its easy to transport,autonomy,and economy.Therefore,this topic focuses on the problem of searching for the wrecked target,and carries out the research on the method and strategy of multi-UUVs task planning and allocation.The main research contents of this article are:Firstly,analyze the task planning of multi-UUVs searching for underwater targets.According to the signal characteristics of the underwater black box and actual engineering requirements,the underwater target search operation is divided into two major contents:finding the coordinates of the black box position and covering and detecting the wreckage of underwater wrecked targets,establish the corresponding mathematical models of the mission environment,underwater target acoustic signals,and UUV active and passive sonar equipment involved in UUV search operations.Secondly,aiming at the requirements of underwater black box search operations,a mission planning strategy for multi-UUV searching underwater targets based on improved neural network model is proposed.By rasterizing the UUV task area,cells containing environmental information are obtained,and then neurons in the neural network model are corresponding to the cells.According to the actual engineering needs and the UUV movement process,assign the UUV movement cost function to the cell.Through data fusion technology,the cell environment information and the motion cost function are fused into a reliability function,and the reliability function value is used as the activity value of the neuron.UUV selects the target location at the next moment according to the neuron activity value around the location,makes UUV search operation less energy consumption.The above method is verified by simulation test.Thirdly,aiming at the detection of underwater wrecked target debris coverage,a task allocation algorithm based on the principle of balance and a path planning strategy for UUV full coverage detection of underwater targets suitable for near seabed environments are designed.The core detection area is partitioned by the Voronori principle,and the UUV task load is evenly distributed based on the principle of biological individual balance.According to the actual analysis of the UUV’s near seabed environment for full coverage detection operations,combined with the intelligent cell reliability function,the UUV area coverage detection underwater target path planning strategy is designed.Finally,based on the Qt development platform,the establishment of a simulation system for multi-UUV searching underwater targets was completed.A simulation test of multi-UUVs searching for underwater targets on this system,and the test results show that the task planning strategy of multi-UUVs searching underwater targets based on the improved neural network model and the coverage strategy based on improved cells are effective and feasible.