Hybrid integral sliding mode and fuzzy logic control for omnidirectional robots: modified elephant herding...

Area of Science:

• Robotics and Control Systems
• Artificial Intelligence
• Optimization Algorithms

Background:

• Autonomous robotic systems face challenges in trajectory tracking and motion control under nonlinear and uncertain conditions.
• Existing control methods may struggle with precision and efficiency in dynamic environments.
• Robust motion control is crucial for reliable operation of mobile robots.

Purpose of the Study:

• To propose a novel hybrid control framework integrating Integral Sliding Mode Control (ISMC), Fuzzy Logic Control (FLC), and Modified Elephant Herding Optimization (MEHO).
• To enhance the trajectory tracking and robust motion control capabilities of a three-wheeled omnidirectional mobile robot (TOMR).
• To evaluate the performance and efficiency of the proposed control strategy compared to existing methods.

Main Methods:

• Implementation of a hybrid control framework combining ISMC, Sugeno-type FLC, and MEHO algorithm for parameter calculation.
• Dynamic and kinematic modeling of a three-wheeled omnidirectional mobile robot (TOMR).
• Utilizing the MEHO algorithm with adaptive mechanisms to optimize exploration-exploitation balance and convergence speed.

Main Results:

• The proposed controller significantly reduced positional errors (X and Y axes) to below 0.005 m and orientation error to 0.0014 rad within 2 seconds.
• Achieved low Root Mean Square Errors (RMSE) for X, Y, and orientation across triangle and C-shape trajectories.
• Demonstrated up to 50% lower torque variation and over 60% faster settling time compared to classical EHO-based and adaptive neural sliding controllers.

Conclusions:

• The novel hybrid control framework offers superior performance in trajectory tracking and robust motion control for autonomous robots.
• The MEHO algorithm effectively optimizes the control parameters, leading to enhanced system accuracy and efficiency.
• The modular and learning-based design shows potential for generalization to other robotic platforms operating in complex environments.