Optimized Power Rate Sliding Mode Control for a Robot Manipulator Using Genetic Algorithms

Omar Mohamed Gad,Raouf Fareh,Hissam Tawfik,Saif Sinan,Sofiane Khadraoui,Maamar Bettayeb
DOI: https://doi.org/10.1007/s12555-024-0187-5
IF: 2.964
2024-10-06
International Journal of Control Automation and Systems
Abstract:This research introduces an efficient power rate sliding mode control (PR-SMC) design methodology to control a 4-degree-of-freedom (DOF) manipulator in both joint and workspace areas. The proposed sliding mode control (SMC) strategy focuses on improving the robot manipulator's tracking accuracy by dynamically fine-tuning the control parameters via the genetic algorithms (GA) optimization approach. To achieve the required joint angles, inverse kinematics of the robot are used. In contrast, forward kinematics are applied to ascertain the precise Cartesian position and orientation of the manipulator's end effector based on the actual joint angles. The genetic algorithm is used to find suitable values of the SMC parameters to ultimately achieve the desired performance of robotic tasks. Additionally, the improved sliding mode control law, integrated with the optimization technique, is practically tested on a 4-DOF manipulator to showcase the practicality and effectiveness of the proposed controller. Finally, the efficacy of the proposed control method is demonstrated by comparing its performance with the performances of a manually tuned PR-SMC and the optimized PR-SMC using particle swarm optimization (PSO). This comparison demonstrates the favorable performance of the proposed method. The experimental results showed that the optimized SMC using the GA outperforms the one optimized using PSO, as it provides less tracking error. The manually tuned SMC, which relies on a trial-and-error method, exhibits a larger error than the other two approaches. Although the PSO algorithm is faster than the GA, the GA showed better accuracy and fitting performance.
automation & control systems
What problem does this paper attempt to address?