Neural Network based PID Control for Two-Axis Inertial Stabilized Platform

This paper presents a two-axis Inertial Stabilized Platform (ISP) equipped with Permanent Magnet Synchronous Motors (PMSM). A stabilizer algorithm utilizing a PID controller with a neural network-based parameter updating method is proposed. The use of PMSM motors in the stabilized system reflects a current technological trend due to their compact size, high torque, gearless design, minimal noise, and low maintenance requirements. Additionally, artificial neural networks can approximate any nonlinear function, allowing for the development of a method to update a set of PID parameters and enhance the adaptability of nonlinear systems across different operating ranges. Experimental results show a significant reduction in tracking error, confirming the effectiveness and adaptability of the proposed neural network-based PID parameter updating method in maintaining system stability across diverse conditions.