| Maritime unmanned aerial vehicles represented by multi-rotor UAVs have the characteristics of vertical take-off and landing,light and flexible,and can carry loads,which are ideal for performing tasks such as maritime monitoring,search and rescue,and sea patrol.However,the flight time of multi-rotor UAVs is relatively short,with only 15-20 minutes to perform the task.Due to the high cost of fixed charging piles and the irregular mission area,it is usually difficult to ensure effective endurance when UAVs are operating at sea.When the UAVs run low on fuel,they coordinate with the USV to find their respective rendezvous points,land on the USV,recharge automatically,and continue their missions under the condition that the overall recharging distance of the UAV cluster is optimal.This thesis combines a mixed integer linear programming approach to such UAVunmanned surface ship cluster autonomous cooperative heterogeneous route planning problems in various aspects and conducts a series of simulation experiments,which can be summarized as follows:(1)The single unmanned surface ship route planning problem and the unmanned surface ship formation route planning problem under mission constraints with load limits and time windows are solved into a graph theoretic problem,and a mixed integer linear programming method is used to model this joint problem,a two-level plug-in method and a meta-heuristic method combining taboo search are given,and comparative experiments are carried out on a standard test set with unmanned surface ship formation simulation experiments to obtain the optimal solution satisfying all conditions.(2)A three-level modeling method that decouples UAV route planning from unmanned surface ship route planning is proposed to statute the problem of solving the UAV and unmanned surface ship route planning under mission constraints into solving the traveling salesman and complex vehicle path planning problems for planning the routes of unmanned surface ships and UAV clusters to achieve coordinated route planning.The objective of this method is to require the UAV cluster to minimize the distance traveled by the UAV cluster under finite energy constraints,time window constraints,and speed constraints to achieve system-wide optimal path planning.(3)Simulation experiments are carried out in 480 different scenarios for verification.The simulation results show that compared with scenarios with scattered task points,scenarios with clustered task points have a higher probability of finding feasible solutions within the time limit,and increasing the number of UAVs can improve the probability of finding feasible solutions,and increasing the number of clusters can make UAVs find more suitable charging stations,which verifies the universality of the proposed method for different tasks and the good synergy between UAVs and unmanned surface ships. |
