Can an intelligent society of robots survive in a hostile environment?

In this paper, we introduce a distributed intelligent system capable of adapting to a possibly hostile environment, by learning and applying task solution strategies as a society. The system is composed of intelligent architectures, that interact with each other, to resolve a common objective. The intelligent entities, themselves, are robotic platforms, equipped with static (microcontrollers/microprocessors) and dynamic (FPGAs) computational resources, proprio-(internal) and extero-ceptor (external) capabilities, as well as various on-board actuators. The result is a society of mobile and stationary intelligent architectures, that are able to utilize a hardware-based genetic algorithm (GA) as an evolutionary computational paradigm, and reconfigurable computing approaches to handle the real-time issues in the entire system. The presented work is part of the ongoing research in the concept of intelligent systems and agents, first presented by Rami Abielmona and Voicu Groza in (2001) and (2002). The distributed intelligent system provides four main services, namely the communications, tracking, monitoring and exploring services. Along with these services, three levels of intelligence will be discussed: local, network and coupled intelligence. Local intelligence describes the know-how of each architecture, network intelligence describes the know-how of cliques of architectures, while coupled intelligence describes the know-how of architectures currently working together. It is envisioned that the architectures will be able to develop human-like instincts, while competing for limited resources in the environment that they reside in.