On-line Path Planning Of USV Under Dynamic Threat In Considering The Ocean Environment

Unmanned Surface Vehicle is a robot or boat with robotic technology,and most cases are arranged in some of complex waters to perform special tasks such as probing or monitoring.Unmanned Surface Vehicle often experience dynamic threats when navigating in these complex seas.In that moment they need to estimate dynamic state of movement and make online route planning combine with their tasks which can adjust their state of motion to ensure that the ship can avoid danger.At the same time,Unmanned Surface Vehicle also need to consider the impact of the marine environment to ensure the accuracy of online route planning.Therefore,it is necessary to study the online route planning method of the unmanned surface vehicle based on the dynamic threat while considering the marine environment.Based on the analysis of Bayesian theory and marine environment-related models,this paper uses Bayesian estimates to predict the state of dynamic threats and Bayesian network reasoning to assess the threat of unmanned surface vehicles.Further,consider the marine environmental impact based on these estimates and evaluations,an on-the-road Unmanned Surface Vehicle route planning method was designed.Firstly,the Bayesian network reasoning and Bayesian estimation are introduced and analyzed in detail,and the motion model of the surface movement are analyzed.The motion control model of the Unmanned Surface Vehicle and the motion model of dynamic threat are deduced.The dynamic model of dynamic is deduced according to the observation information.The dynamic model of Unmanned Surface Vehicle is also analyzed based on rigid body motion and moment model.Secondly,according to the motion model and observation model which are derived to design a state estimation method based on double Bayesian estimation.First,the threat velocity and direction are calculated from the observed value,and then the first-level Bayesian estimation is obtained the velocity and direction at this time.the dynamic threat position and the first-level Bayesian estimation result are input into the second-level Bayesian estimation,and the dynamic threat position estimation is obtained.While using this estimation method to estimate the motion state of the dynamic threat when consider the marine environment.This method can grasp the movement model better and be more accurate than the traditional estimates.Again,innovatively use the Bayesian network reasoning to deal with this problem under the unintellectual and uncontrollable of dynamic threat.The motion state estimation based on the known dynamic threat is used to design a fuzzy discrete dynamic Bayesian and according to this to assess the threat probability of the Unmanned Surface Vehicle.Threat probability information is used as the input of the online route planning.Finally,based on the idea of forecasting rolling optimization in forecast model,the motion state estimation of dynamic threat and the threat probability assessment information of unmanned aerial vehicle are taken as input.A method of forecasting rolling optimization of online route planning of Unmanned Surface Vehicle is proposed.The method of simulation and optimization of the forecasting rolling optimization route is verified by considering the marine environment and unconsidering the marine environment.The simulation results shows that the predictive rolling optimization route planning in this paper can make the Unmanned Surface Vehicle effectively avoid the dynamic threat to perform the task.