AUTHOR=Jasim Mohamed Mohamed , Oleiwi Bashra Kadhim , Azar Ahmad Taher , Mahlous Ahmed Redha 

TITLE=Hybrid controller with neural network PID/FOPID operations for two-link rigid robot manipulator based on the zebra optimization algorithm

JOURNAL=Frontiers in Robotics and AI

VOLUME=Volume 11 - 2024

YEAR=2024

URL=https://www.frontiersin.org/journals/robotics-and-ai/articles/10.3389/frobt.2024.1386968

DOI=10.3389/frobt.2024.1386968

ISSN=2296-9144

ABSTRACT=This paper proposes six control structures based on Neural Network (NN) with Proportional Integral Derivative (PID) controller and Fractional Order PID controller (FOPID) to control a 2-Link Rigid Robot Manipulator (2-LRRM) for trajectory tracking issue. These proposed controllers are named set point Weight Proportional Integral Derivative (W-PID) controller, set point Weight Fractional Order PID (W-FOPID) controller, Recurrent Neural Network like PID (RNN-PID) controller, Recurrent Neural Network like FOPID (RNN-FOPID) controller, Neural Network plus PID (NN+PID) controller and Neural Network plus FOPID (NN+FOPID) controller. The Zebra Optimization Algorithm (ZOA)utilizes in order to adjust the parameters of the proposed controllers while reducing the Integral Time Square Error (ITSE). A new objective function proposes for tuning process in order to generate a controller with the least amount of chattering in the control signal. After implementing the proposed controller designs, a comparative robustness study was conducted among the suggested controllers by altering the initial conditions, disturbances, and uncertainties of the model. The simulation results demonstrate that the NN+FOPID controller performs the best trajectory tracking with the minimum ITSE and also has the best robustness against the changes in the initial states, external disturbances, and parameter uncertainties as compared with other proposed controllers.