Agent-based artificial immune system model for the detection of faults in a distributed satellite system

A distributed satellite system is composed of several small satellites working collaboratively or symbiotically together on a satellite mission. In such a system, fault detection can be executed in a similar way the distributed and specialized cells of the biological immune system find and counteract sicknesses. This paper proposes the use of an agent-based artificial immune system model for the detection of faults. The method takes into account the reduced memory and computational resources that small satellites have, applying control of population of agents through death and reproduction, and dimension reduction of the signatures that identify the agents. Simulations of the dynamics of the population of agents, of a model adjusted to a system with several satellites, shows emergent properties for encountering faults and counteracting them correspondingly.