Autonomic systems are typically distributed, complex and concurrent systems, comprised of multiple interacting autonomic elements that often exhibit emergent behavior. Design and development of such systems is a non-trivial task that by definition requires specific software engineering approaches, including the use of specialized modeling techniques. This paper takes the well-known Prospecting Asteroids Mission from NASA, an autonomous and autonomic system for future exploration of asteroid fields, which can be considered as a axiomatic example of how autonomic principles can be applied in real terms. Introducing the Agent Modeling Language (AML), we demonstrate using a series of didactic examples how AML can be applied to efficiently, accurately and comprehensively model the PAM system. A selection of AML models specifying the PAM domain, goals, architecture, and behaviors are presented which help demonstrate the utility of AML when modeling autonomous and autonomic systems.
Citation:
Radovan Cervenka, Dominic Greenwood, Ivan Trencansky, "The AML Approach to Modeling Autonomic Systems," icas, pp.29, International Conference on Autonomic and Autonomous Systems (ICAS'06), 2006