Exploiting systemic biological modeling for trigger based adaptation in networked intelligent multi-agent systems

Current day networked intelligent agent based systems have limited capability of adaptability, self-repair, adaptation, and self-reconfiguration under changing external conditions. In past, evolutionary algorithms have experimented with random mutation and heuristic selection based evolution for self-adaptation. However, little research has been done to explore dynamic adaptive control to take care of sudden external stress and events at systemic response level. This work introduces a new message based biological model of intelligent multiagent based systems that represents agents as self-correcting dynamically modifiable genes - a reconfigurable set of dynamically regulated built-in functions, and system of agents as dynamically adaptable event-trigger controlled interacting pathways that can be altered and reconfigured in response to external stress and events. The model supports the integration of message, code, trigger, and belief states, and supports interchangeability of message, code, and trigger to provide dynamic adaptive control. The model and its implementation have been described.