Design and Implementation of Tuning PID Controller using Modern Optimization Techniques for Micro-Robotics System
Ehab S. Ghith1, Mohamed Sallam2, Sherif A. Hammad3
1Ehab S. Ghith *, A Ph.D. Candidate, Department of Mechatronics, Faculty of Engineering, Ain shams University, Cairo, Egypt.
2Mohamed Sallam, Assistant Professor of Mechatronics Engineering, Helwan University, Cairo, Egypt.
3Sherif A. Hammad, Professor at Mechatronics, Faculty of Engineering, Ain shams University, Cairo, Egypt.
Manuscript received on August 25, 2021. | Revised Manuscript received on August 30, 2021. | Manuscript published on September 30, 2021. | PP: 51-68 | Volume-10 Issue-11, September 2021. | Retrieval Number: 100.1/ijitee.J945408101021 | DOI: 10.35940/ijitee.J9454.09101121
Open Access | Ethics and  Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Abstract: One of the main difficult tasks in the field of micro-robotics is the process of the selection of the optimal parameters for the PID controllers. Some methods existed to solve this task and the common method used was the Ziegler and Nichols. The former method require an accurate mathematical model. This method is beneficial in linear systems, however, if the system becomes more complex or non-linear the method cannot produce accurate values to the parameters of the system. A solution proposed for this problem recently is the application of optimization techniques. There are various optimization techniques can be used to solve various optimization problems. In this paper, several optimization methods are applied to compute the optimal parameter of PID controllers. These methods are flower pollination algorithm (FPA), grey wolf optimization (GWO), sin cosine algorithm (SCA), slime mould algorithm (SMA), and sparrow search algorithm (SSA). The fitness function applied in the former optimization techniques is the integral square Time multiplied square Error (ISTES) as the performance index measure. The fitness function provides minimal rise time, minimal settling time, fast response, and no overshoot, Steady state error equal to zero, a very low transient response and a non-oscillating steady state response with excellent stabilization. The effectiveness of the proposed SSA-based controller was verified by comparisons made with FPA, GWO, SCA, SMA controllers in terms of time and frequency response. Each control technique will be applied to the identified model (simulation results) using MATLAB Simulink and the laboratory setup (experimental results) using LABVIEW software. Finally, the SSA showed the highest performance in time and frequency responses.
Keywords: Flower pollination algorithm, PID Controller, Micro-robotics, Grey wolf optimization, Sine Cosine algorithm, slime mould algorithm, Sparrow search algorithm.