Abstract:

A constraint satisfaction model and a discrete differential evolution combined with variable neighborhood search (DDE-VNS) are proposed in this paper for an agile satellite task scheduling considering the influence of the observation time on the imaging quality. Firstly, the temporal constraint of the agile satellite task scheduling is described. Secondly, considering the influence of the observation time on the imaging quality, the constraint of the attitude transition between tasks, and onboard storage and energy, the constraint satisfaction model of the agile satellite task scheduling is established. Thirdly, the mutation, crossover and selection operators of the discrete differential evolution are designed, the variable neighborhood search is adopted to exploit better solution from the optimal solution at each iteration, and also the workflow of the proposed algorithm is described. Simulations have demonstrated that the model and algorithm proposed can get schedule plans with cracking reward and quick converge.

Key words: Agile satellite, Task scheduling, Discrete differential evolution, Variable neighborhood search