APPLICATION OF FUZZY LOGIC IN THE ROBOT CONTROL FOR MECHANICAL PROCESSING

Robot application in mechanical machining is growing day by day because it has many advantages over conventional machines such as high flexibility, large working space, and high repeatability. Many degrees of freedom of motion give robots the ability to perform complex technological operations, but also because of that, methods of controlling robots based on dynamic models have difficulties. Applying fuzzy logic to robot control can partially or completely exclude the calculation of the robot's dynamic model as well as overcome other uncertainties of the whole technological system. The variables and parameters of the fuzzy logic-based controller are modeled in a linguistic form, called linguistic variables, and are defined by the linguistic semantic values. Fuzzy rules are an important basis for the performance of operations defining the control quantities of the controller. Fuzzy rules are constructed by natural human inference and are based on expert intelligence. The main tasks of applying fuzzy logic control include “Fuzzification” to determine fuzzy parameters in the form of fuzzy sets of input-output data; “Fuzzy Rules and Fuzzy Inference Mechanism” to perform fuzzy operations defining control quantities, and finally “Defuzzification” to convert control quantities from linguistic values to physical values for controller operation. There have been many types of research on applying fuzzy logic to control robots in general, but the percentage of fuzzy control research work for mechanical machining robots is still limited. The article is based on published works on fuzzy control for robots in general and mechanical processing robots to analyze the applicability of fuzzy control for mechanical machining robots. The article provides detailed information on fuzzy controller design, on determining input and output variables, proportional mapping to determine the number of fuzzy sets and the corresponding type of membership function. The construction of fuzzy rule base system and fuzzy inference mechanism is presented, and finally defuzzification. Prospects for the use of methods to perfect and develop fuzzy control systems for mechanical machining robots are also presented. Collectively, the information in this document is intended to guide the implementation of fuzzy logic-based controller designs for application to machining robots, as well as to general robot control.