无人机多任务区侦察时间分配模型与应用

为了合理分配无人机对多个任务区的侦察时间，提出了一种包含问题建模、求解和方案决策的无人机多任务区侦察时间分配方法。首先，建立了包含侦察收益和侦察风险两目标的无人机多任务区侦察时间分配模型，该模型属于带约束多目标优化问题；其次，提出了一种改进的基于分解的约束多目标进化算法，该算法具有简单、灵活、无参等特点，可有效求解；最后，利用优劣解距离法从非支配解集中选择最优方案。选择了6种约束多目标进化算法，在3个不同雷达强度指数条件下进行对比实验。Hypervolume指标说明约束多目标进化算法在求解该问题时优于其他算法。实验结果表明：提出的方法在求解无人机多任务区侦察时间分配优化问题中能够实现快速准确决策。

Model and Application on UAV Multi-task Area Reconnaissance Time Allocation

In order to reasonably allocate the reconnaissance time of unmanned aerial vehicle (UAV) in multi-task areas, an UAV multi-task area reconnaissance time allocation method is proposed in this paper, which includes establishing problem model, solving and scheme decision. First, a UAV multi-task area reconnaissance time allocation model including reconnaissance gains and reconnaissance risks is established. This model belongs to a multi-objective optimization problem with constraints. Second, an improved constrained multi-objective evolutionary algorithm based on decomposition is proposed to solve the problem. The algorithm is simple, flexible and nonparametric, so that it can solve constrained multi-objective optimization problems quickly. Finally, a multi-criteria decision-making method is used to select the optimal scheme from the non-dominated solution set. Six constrained multi-objective evolutionary algorithms were selected for comparative experiments under three different radar intensity indexes. The hypervolume index shows that the proposed constrained multi-objective evolutionary algorithm in this paper is superior to the other algorithms in solving this problem. The experimental results show that the proposed method can realize fast and accurate decision-making in optimizing UAV multi-task area reconnaissance time allocation.