On-board Dynamic Scheduling with a Fuzzy Logic Inference Engine

This paper presents a new approach to face the autonomous updating of the on board functions for space applications. The problem is treated as a multi-criteria decision making on a dynamic knowledge base of the e ntire system. A search direction technique is proposed b ased o n different ranking processes on the possible decisions in solving v iolations. The heuristics that orders and prunes a list of choices is iteratively created through dedicated fuzzy blocks. The fuzzy logic has been selected to simulate the human behavior in making decision within complex domains. The activities re-allocation is implemented through the predicate calculus on the knowledge base. The proposed method gives a final scheduling scenario at a very high level. The European Space Agency PROBA mission has been selected to validate the scheduler.