NN Based Active Disturbance Rejection Controller for a Multi-Axis Gimbal System

TL;DR

This paper introduces neural network assisted active disturbance rejection control (ADRC) for multi-axis gimbal systems, significantly improving tracking accuracy and robustness by reducing mean errors through innovative disturbance compensation methods.

Contribution

It presents a novel neural network assisted ADRC approach that enhances disturbance rejection and reduces tuning effort in multi-axis gimbal control systems.

Findings

NN assisted ADRC reduces mean tracking errors by 85.4%.

CTM assisted ADRC reduces errors by 40.8%.

Both methods improve robustness and environmental adaptability.

Abstract

The increasing demand for target tracking, environmental surveys, surveillance and mapping requires multi-axis gimbal systems with high tracking and stabilization performance. In this paper, first, computed torque model is generated to estimate the complex disturbances acting on the system. Then, two different control strategies based on active disturbance rejection control (ADRC) and computed torque model are implemented on a two-axis gimbal system. The purpose is to improve the robustness, environmental adaptability and tracking accuracy of the system and reduce the tuning effort of ADRC by integrating a neural network (NN) based disturbance compensator (NN assisted ADRC). In the second control strategy, NN is replaced with a computed torque model (CTM assisted ADRC), whose inputs come from plant outputs. The simulation results show that, NN and CTM assisted ADRC structures can decrease mean tracking errors up to 85.4% and 40.8%, respectively.