Cognitive NFIDC-FRBFNN Control Architecture for Robust Path Tracking of Mobile Service Robots in Hospital Settings

Huda Talib Najm^1,2, Ahmed Sabah Al-Araji³, Nur Syazreen Ahmad^1,*
1 School of Electrical and Electronic Engineering, Universiti Sains Malaysia, Nibong Tebal, 14300, Penang, Malaysia
2 Biomedical Engineering Department, University of Technology, Baghdad, 10066, Iraq
3 Control and Systems Engineering Department, University of Technology, Baghdad, 10066, Iraq
* Corresponding Author: Nur Syazreen Ahmad. Email: email

Computer Modeling in Engineering & Sciences https://doi.org/10.32604/cmes.2025.071837

Received 13 August 2025; Accepted 10 November 2025; Published online 04 January 2026

Download PDF

Abstract

Mobile service robots (MSRs) in hospital environments require precise and robust trajectory tracking to ensure reliable operation under dynamic conditions, including model uncertainties and external disturbances. This study presents a cognitive control strategy that integrates a Numerical Feedforward Inverse Dynamic Controller (NFIDC) with a Feedback Radial Basis Function Neural Network (FRBFNN). The robot’s mechanical structure was designed in SolidWorks 2022 SP2.0 and validated under operational loads using finite element analysis in ANSYS 2022 R1. The NFIDC-FRBFNN framework merges proactive inverse dynamic compensation with adaptive neural learning to achieve smooth torque responses and accurate motion control. A two-stage simulation evaluation was conducted. In the first stage, the controller was tested in a simulated hospital environment under both ideal and non-ideal conditions. In the second, it was benchmarked against four established controllers - Neural Network Model Reference Adaptive (NNMRA), Z-number Fuzzy Logic (Z-FL), Adaptive Dynamic Controller (ADC), and Fuzzy Logic-PID (FL-PID)—using circular and lemniscate trajectories. Across ten runs, the proposed controller achieved the lowest tracking errors under all conditions. Under ideal conditions, it achieved average improvements of 55.24%, 75.75%, and 55.20% in integral absolute error (IAE), integral squared error (ISE), and mean absolute error (MAE), respectively, with coefficient of variation (CV) reductions above 55%. Under non-ideal conditions, average improvements exceeded 64% in IAE, 77% in ISE, and 66% in MAE, while maintaining CV reductions above 57%. These results confirm that the NFIDC-FRBFNN controller offers superior accuracy, robustness, and consistency for real-time path tracking in healthcare robotics.