Proceedings of the 6th International Symposium on Artificial Intelligence, Robotics and Automation in Space: A New Space Odyssey, Montreal (June 2001)
In this paper, we present a system for Distributed Intelligent Planning and Scheduling (DIPS) that helps a spacecraft function as an autonomous agent. A DIPS-based spacecraft receives only high-level goals from ground station operators, and performs its own planning and scheduling onboard, achieving these goals with onboard subsystems and in cooperation with other spacecraft. The task decentralization in DIPS employs a domain distribution algorithm that typically creates a feasible schedule after the first coordination effort, thereby decreasing inter-agent negotiation during the scheduling process. The reasoning performed by DIPS agents to optimize time and resource usage while maintaining flight rules and constraints is based on a constraint propagation paradigm. Priority-based scheduling is implemented, and a hierarchical inter-agent confirmation/authorization system is used for global goal coordination. An enhanced prototype is developed and demonstrated using space-based scenarios involving onboard instruments and a satellite constellation. The vertically layered architecture of the DIPS prototype integrates: 1) Java-based agent inference engine; 2) Prolog platform SICStus for constraint-based reasoning; and 3) KQML for inter-agent communication. We are specifically targeting our effort to enhance the planning and scheduling capability of NASA’s planned nanosatellite constellations.
For More Information
(Please include your name, address, organization, and the paper reference. Requests without this information will not be honored.)