AI Based Tools for Continuous Support to Mission Planning

A. Cesta, A. Oddi, G. Cortellessa, S. Fratini, and N. Policella

In Proceedings of the 9th International Conference on Space Operations (SpaceOps 06), 2006

This paper shows how Artificial Intelligence (AI) techniques for planning, scheduling, domain modeling, and intelligent interaction have been applied for solving a challenging problem: supporting mission planning activities for space applications. We propose a generic software architecture for solving complex mission planning problems composed by two modules. A first one, called the Problem Solver (PS), which models the relevant knowl- edge of the problem as a Constraint Satisfaction Problem (CSP) and uses a set of automatic algorithms for generating problem solutions on top of this representation. The second component of the system is the Man-Machine Interface (MMI), that directly interacts with the user, and allows her/him to keep control at some level on the automated solver operations. The work gives a twofold contribution. On one hand, it propose the Constraint Satisfaction Problem (CSP) as a reference model for representing mission planning problems and presents a set of metaheuristic strategies based on the CSP model of the problem. On the other hand, it remarks as the interaction with the user is a key aspect for the success of technological tools and proposes few and effective principles for exploiting human capabilities within the solving process of a complex mission planning problem. Starting from the O-Oscar tool - the outcome of a project work funded by the Italian Space Agency (ASI) - we shows how this core architecture is evolved in solving a complex mission planning problem related to the Mars-Express program of the European Space Agency (ESA).We describe a two-step effort: a first study, which produced a demonstration prototype able to capture the main aspects of the problem, and a second effort, that has developed a fielded application, completely integrated in the operations environment. The tool is currently used on daily basis by the mission planning team of Mars-Express at the European Space Agency since February 2005.