This paper uses eight evaluation metrics to compare and contrast three coordination schemes for a system that continuously plans to control collections of rovers (or spacecraft) using collective mission goals instead of goals or command sequences for each spacecraft. These schemes use a central coordinator to either (1) micromanage rovers one activity at a time, (2) assign mission goals to rovers, or (3) arbitrate mission goal auctions among rovers. A self-commanding collection of rovers would autonomously coordinate itself to satisfy high-level science and engineering goals in a changing partially understood environment-making the operation of tens or even a hundred spacecraft feasible.
展开▼