A paper presents the concept of a system for autonomous communication among multiple satellites and other spacecraft. The design of the system would be based on an evolvable architecture of multiple intelligent agents (that is, artificial-intelligence constructs implemented in software and hardware) that would communicate and cooperate with each other in performing such tasks as enabling each spacecraft to track the others, analyzing communication links, dynamically making and breaking links, and otherwise generally allocating communication resources. The paper discusses some of the problems to be addressed in constructing the system, describes an example of an evolvable system containing knowledge-based agents, and describes some hardware modules that could be used in building the system.
This work was done by Sanda Mandutianu of Caltech for NASA's Jet Propulsion Laboratory. NPO-20432
This Brief includes a Technical Support Package (TSP).
Evolvable multiagent approach to spacecraft communication
(reference NPO20432) is currently available for download from the TSP library.
Don't have an account?
Overview
The document presents an innovative approach to spacecraft communication through an evolvable multi-agent system developed by Sanda Mandutianu at NASA's Jet Propulsion Laboratory. The system is designed to enable autonomous communication among multiple satellites and spacecraft, utilizing a hybrid architecture of intelligent agents that can communicate and cooperate to perform various tasks. These tasks include tracking each other, analyzing communication links, and dynamically managing communication resources.
The core concept revolves around the use of evolutionary algorithms to develop artificial intelligence rules that adapt to changing circumstances, enhancing the system's operational robustness. The agents are capable of evolving, which allows them to respond effectively to the dynamic environment of space communication. This adaptability is crucial for maintaining efficient communication links and resource allocation among spacecraft.
The document also discusses the hardware implementation of these rule-based agents. It highlights the development of a low-power, high-speed analog programmable chip designed specifically for implementing fuzzy rule-based systems. This chip can process multiple input variables and execute fuzzy logic operations, enabling rapid decision-making in the context of spacecraft communication. The chip is capable of handling 16 input variables and providing 32 decision outputs, with a decision latency of approximately one microsecond.
Additionally, the paper outlines a potential solution for improving data transmission efficiency. It suggests evolving algorithms for data compression and defining new agents with enhanced strategies to alleviate the burden on existing data transmission agents. This approach aims to optimize the performance of the communication system, ensuring timely and efficient data transmission.
The document concludes by emphasizing the importance of this evolvable multi-agent approach in addressing the challenges of spacecraft communication. By integrating advanced artificial intelligence and hardware evolution methods, the proposed system aims to create a more adaptive and robust communication network for future space missions.
Overall, the paper provides a comprehensive overview of the proposed system, its underlying technologies, and the potential benefits of implementing an evolvable multi-agent architecture in spacecraft communication, paving the way for more efficient and autonomous operations in space.
