Research On Path Planning Of An UAV In Water Environment Monitoring

As is known, the flight path of an unmanned aerial vehicle (UAV) directly affects its efficiency. However, due to the limitations of flight performance, an UAV is unable to cover all the areas at one time when monitoring in the water environment. Besides, different monitoring tasks have different requirements for the path planning. For example, an UAV, when monitoring a target on regular patrols, demands the total time as long as possible, and would immediately switch to the feasible path or the second-best path once the target is altered. Also, compared with flying in the terrestrial environment, there is a greater risk navigating in the water environment for the UAV. Traditional path planning provides the flight path points by human and the UAV flying according to these points which leads to inefficiency of monitoring performance, or even safety problem. Therefore, the path planning in the water environment is of great significance for research. This paper, with a focus on the UAV used for monitoring in the water environment, aims to optimize the flight path according to the specific requirements for different monitoring tasks. The main contents of this paper include:Studying the application of a UAV monitoring in the water environment, reviewing the present researches on path planning and commonly-used algorithms, specifying the requirements for path planning in the water environment, and giving description on the primary characteristics and the processes of both genetic algorithms (GA) and simulated annealing (SA) algorithm;Establishing a mathematical model for path planning by the UAV according to the specific requirements for the water environment monitoring. Introducing the hierarchical thinking of path planning, then giving a mathematical description on path planning and a study on the track constraints, and setting up the objective function according to the mission requirements;Studying the overall path planning of the UAV when monitoring in the water environment. As the genetic algorithm is susceptible to the local optimal solution, employing a genetic-simulated annealing algorithm. Solving the problem of the minimum-step size constraint by using a polar coordinate encoding method, then working out the solution based on the genetic-simulated annealing algorithm whose initial solution is input by the genetic algorithms. The simulation outcomes suggest that a better path, also the optimum path as discussed in the paper, can be achieved by employing the genetic-simulated annealing algorithm;Studying the online route of the UAV that monitors in the water environment, proposing an online path planning algorithm based on the genetic algorithm. Examining the temporal complexity of computing the online route planning based on the solutions produced by the online path planning method, during which the optimum path is obtained as the reference path in the overall path planning. This process also shows that this method can generate feasible path or even second-best path in a very short time.