Towards a Self-Configuring Optimization System for Spacecraft Design

Spacecraft design optimization is a difficult problem, due to the complexity of optimization cost surfaces and the human expertise in optimization that is necessary in order to achieve good results. In this paper, we propose the use of a set of generic, metaheuristic optimization algorithms (e.g., genetic algorithms, simulated annealing), which is configured for a particular optimization problem by an adaptive problem solver based on artificial intelligence and machine learning techniques. We describe work in progress on OASIS, a selfconfiguring optimization system based on these principles.