Multi-layered inference with possibility distribution for multiple micro-machines control

This paper presents the control architecture of multiple micro-machines based on a multilayered inference system within a framework of an intelligent control support system similar to human decision making process. The process can be divided into two phases: the state recognition through analysis of various data, and the prediction of this result. The aim of this study is to construct these processes on a workstation with the following architecture: 1) various numeric values are represented by possibility distribution; and 2) the system is made up of three layers and the abstractness is increased with the increment of the layer levels. Discussions on both the configuration and realization of this system are given