Processor-In-the-Loop Simulation of a Neuro-Fuzzy Controller on Raspberry Pi 3 board

Abstract

In this paper, the study of the ANFIS (Adaptive Neuro-Fuzzy Inference System) controller and its implementation on a Raspberry Pi 3 board were discussed. The objective is to study the behavior of the ANFIS regulator on the hardware target (Raspberry Pi 3 board) in the case of the control in position of an inverted pendulum, using the Processor-In-the-Loop (PIL) technique from MATLAB-Simulink. The proposed approach consists of two stages. The first step concerns the synthesis of the fuzzy neural controller, and in the second step, we present the control scheme synthesized on the Simulink environment, in order to control the position of the inverted pendulum. Finally, the simulation results in PIL mode show the efficiency of the proposed controller in terms of tracking and robustness against disturbances.