Research On AUV Path Planning For Data Completeness In Underwater Sensor Networks

As an effective tool for exploring the ocean,underwater sensor networks(UWSNs)have a wide range of applications in submarine resource detection,marine environment monitoring,marine disaster warning,and other fields.Collecting environmental data perceived by sensor nodes is one of the core tasks of UWSNs.AUV-based data collection methods have brought significant advantages for UWSNs to save and balance energy consumption,so AUV path planning has gradually become a research hotspot.In the marine application of monitoring emergency events,the event data has timeliness.If the event data cannot be collected in time,inaccurate decisions will be made by the maritime application due to incomplete data.At the same time,because of the complex underwater environment and irregular occurrence of events,the data collection requirements of sensor nodes are dynamic.Therefore,according to the data collection requirements of sensor nodes,planning AUV paths on demand is of great significance for improving data completeness and reducing data collection costs.This paper first proposes an AUV on-demand path planning algorithm suitable for static non-flow environments such as reservoirs,lakes,and bays.The goal is to enable AUV to collect more data in time to maximize data completeness.According to the characteristics of sensor nodes with dynamic data collection requirements,the path planning in the algorithm is divided into two parts: path planning of the nodes to be collected and path update.In the specific implementation,firstly,the variable neighborhood search is applied to plan the approximate optimal path of the nodes to be collected to avoid the path planning falling into the local optimal solution.Then,in response to the new requirements,a node insertion algorithm is proposed to update the AUV path to improve the efficiency of data collection.Finally,simulation experiments verify that the algorithm can effectively improve the performance of data completeness.In some environments with current,such as rivers and oceans,current has a nonnegligible effect on the movement trajectory and speed of the AUV.Therefore,based on the previous work,this paper further improves and proposes an AUV path planning algorithm suitable for the dynamic current environment,which integrates the impact of current on AUV into AUV path planning.When only part of the current velocity is known,the algorithm first predicts the global current velocity in the network deployment area based on Kriging interpolation.Then,under the influence of current,the control method of linear trajectory movement and the calculation of movement time between nodes are given.Finally,new problems in path planning under the influence of current are analyzed and corresponding solutions are given.The experimental results show that compared with other data collection methods,this algorithm can effectively improve the performance of AUV data collection and reduce the wrong path planning.