A new control architecture combining reactivity, deliberation and motivation for situated autonomous agent

Intelligent behavior can be observed from both natural and artificial systems, but still the notion of intelligence is very difficult to define. We propose a new control architecture allowing the combination of reactivity, planning, deliberation and motives for building intelligent agents that must deal with real or other kinds of environments suited to their life purposes. This control architecture tries to unify the principles and characteristics associated with intelligence. It uses behaviors as its basic control components. These behaviors are selected dynamically and their actions are combined according to the intentions of the agent. Introspection of its reactions and its knowledge is one major new ability given to the agent by this architecture. This way, the agent is not only able to adapt to the environment, but also to its own capacities. One implementation for experimenting with a simulated environment for mobile robots is presented here to illustrate the use of this architecture.