Detection of Signal Data Alternations by Forward Chaining XPS

  • A. A. Bielskis Klaipėda University
  • O. Ramašauskas Klaipėda University

Abstract

This paper proposes a method of visualisation of robotic devices signal data sensed from ambient environment of non linear system. The proposed model aims to detect the alterations of robot sensors signal data by using newly redesigned Forward Chaining Expert System (FXPS). The proposed simple expert system allows interpreting a matrix type data which describes the motions of robot hand as well as transformations of the objects in a symbolic form. The coherency using Expert System principles with Artificial Neural Networking methods is also discussed. The software implementation in Matlab-based Expert System which provides production rule based signal data processing is also discussed. Visualisation results of the object graphical transformations are presented. Some efforts on support capabilities for the Expert System including the ability to edit rules and update expert knowledge data values have been focused in this paper. It is shown that the flexibility of the Matlab scripting language provides an Expert Systems platform a possibility to evolve if more features of knowledge base and production rules are included. Ill. 4, bibl. 10 (in English; summaries in English, Russian and Lithuanian).

Published
2015-03-16
Section
T 125 AUTOMATION, ROBOTICS