Robustness of Market-Based Task Allocation in a Distributed Satellite System

The perceived robustness of multi-agent systems is claimed to be one of the great benefits of distributed control, but centralised control dominates in space applications. We propose the use of marketbased control to allocate tasks in a distributed satellite system. The use of an artificial currency allows us to take the capabilities, energy levels and location of individual satellites, as well as significant communication costs into account. Simulation is used to compare this approach to centralised allocation. We find the market-based system is more efficient and more robust to satellite failure, due to the adaptive allocation of tasks.