Methodology for the Implementation of a Fuzzy Controller on Arduino, MATLAB™ and Nexys 4™ Platforms

Abstract

This chapter presents a methodology to implement a fuzzy controller in different hardware platforms, which can be used to control a system or process. The methodology proposes a programming algorithm to implement a fuzzy controller on the Arduino UNO, Arduino DUE, Nexys 4™, and MATLAB™ platforms. The programming algorithm uses two control statements (IF-THEN and FOR) and the basic mathematical operations. The fuzzy controller was designed for two input variables, one output variable, five fuzzy sets for each variable, and a Mamdani type structure. An analysis of convergence time, amount of memory, and control surface is performed to ensure that the fuzzy controller on all platforms is satisfactory. MATLAB™ is used to compare these platforms through numerical simulations, which demonstrates the effectiveness of the proposed methodology. The experimental results of the fuzzy controller are a processing time of 117 milliseconds and 40% of the memory of the Arduino UNO, a processing time of 21.275 milliseconds and 5% of the memory of the Arduino DUE, and a processing time of 17.871 milliseconds and 40% of the memory on the Nexys 4™. Finally, a Mean Square Error of 0.0326, 0.0643, and 0.1125 was obtained for MATLAB™, Arduino, and Nexys 4™, respectively.