AUTONOMOUS INTELLIGENT AGENT CONTROL IN COMPLEX ENVIRONMENTS

The paper deals with a problem of intelligent system's design for complex environments. There is discussed a possibility to integrate several technologies into one basic structure that could form a kernel of an intelligent system or an intelligent agent. One alternative structure is proposed in order to form a basis for an intelligent system that would be able to operate in complex environments. The proposed structure is very flexible because of the features that allow adaptation via learning and adjustment of the existing knowledge. Therefore the structure may be used in environments with stochastic features such as hardly predictable events or elements (intelligent entities). This paper discusses a possibility to use the proposed structure in order to model intelligent entities or entire intelligent systems in agent based modeling domains [Davidsson 2000, Drogoul 2002]. The proposed structure has found its implementation in experimental robotic system. The robotic system has been used for experimentations in order to validate the structure as well as its functionality, flexibility and reliability. Features of the robotic system and the most important results of the experiments are outlined and commented at the end of the paper. Some possible directions of further research are also sketched at the end of the paper.