Onboard mission planning for agile satellite using modified mixed-integer linear programming

This paper investigates a new Agile Earth Observation Satellite (AEOS) mission planning algorithm. This new developed algorithm is based on Modified Mixed-Integer Linear Programming (MILP) approach, which takes the minimum slew angle and highest priority criterion. The key point of the paper lies in that the planning process is treated as a Dynamical Combinatorial Optimization (DCO) problem due to the time-varying constraint and the requirement of the problem. The design of the new algorithm is also to meet the real-time requirement of onboard mission planning. The mathematical formulation of the problem is modified to satisfy the demand of the LP solver, and a newly developed dynamic database is used to simulate the constantly changing satellite-target relative position with limited observation windows. The new mission planning model is solved by the LP method to test its reliability. Another solution with Generic Algorithm (GA) is also given to check the result. The calculation time of these two methods is compared to show that the LP approach has a better compatibility with the onboard mission planning requirement.