International Conference on Computer and Knowledge Engineering

Home / 15th International Conference on Computer and Knowledge Engineering

Intelligent Adaptive Control of Wheeled Mobile Robots with MLP Neural Networks under Input Constraints

Authors :

Soheil Sheikh ahmadi¹ Arash Rahmani²

1- Faculty of Electrical and Computer Engineering, University of Tabriz 2- Faculty of Mechncical Engineering, Urmia University of Technology

Keywords :

Intelligent Adaptive Control،Wheeled Mobile Robots،Multi-Layer Perceptron،Trajectory Tracking،Input Constraints

Abstract :

This paper presents an intelligent adaptive control framework for wheeled mobile robots (WMRs), leveraging Multi-Layer Perceptron (MLP) neural networks to address the dual challenges of trajectory tracking and system stabilization under input constraints. The proposed controller incorporates a real-time parameter adaptation mechanism that effectively compensates for model uncertainties and external disturbances, while concurrently handling actuator saturation—a common limitation in robotic platforms. By embedding the learning capabilities of MLPs within the adaptive control structure, the controller continuously updates its control policy to ensure high-precision tracking, even under constrained input conditions. Comprehensive simulation results confirm the effectiveness of the proposed approach, highlighting significant reductions in steady-state error and substantial improvements in trajectory tracking accuracy. The demonstrated robustness and adaptability of the controller render it a strong candidate for real-world deployment in autonomous robotic systems operating in dynamic and uncertain environments.