Skip to content
Open access

PID Tuning Approach Using Kolmogorov–Arnold Networks Under Varying Operational Conditions

Sep 2026 · Applied Sciences · Vol 16, pp. 9597 · 0 citations · 23 references

Abstract

This paper investigates the use of Kolmogorov–Arnold networks (KANs) for dynamic proportional-integral-derivative (PID) control tuning in first- and second-order linear systems under noisy conditions, time-varying reference trajectories, and varying plant parameters. Toy datasets, based on instantaneous system error, output and reference trajectory, are used for training the networks and comparing KAN results with fixed PID coefficients, taken from MATLAB’s Simulink PID Autotune, a multilayer perceptron (MLP)-based neural network (NN), trained on the same datasets, a traditional adaptive PID scheme with gain scheduling and an LMS-based online tuning approach. Results evaluate KAN performance under several scenarios, including time-varying reference trajectories, measurement noise, and changes in the controlled plant. The results show that the learning-based tuning approaches outperform gain scheduling and LMS-based adaptation under non-stationary operating conditions. The KAN-based tuner achieves comparable or improved tracking performance relative to the MLP and consistently provides reduced steady-state and integral error measures compared with the classical approaches. While KAN and MLP controllers exhibit similar performance for first-order plants, the KAN-based tuner demonstrates improved robustness for the more complex second-order systems, particularly during reference transitions and under high measurement-noise conditions. Analysis of the gain trajectories further shows that KANs provide rapid adaptation to changes in system dynamics and operating conditions, although this increased responsiveness can result in greater gain variability under noisy conditions. A parameter-matched comparison further indicates that MLP architectures with the same parameter budgets as the KAN models provide inadequate performance, whereas substantially wider MLPs are required to achieve competitive results.

Read PDF

Similar papers

Open access Sep 2026

Riccati-Based Static Output-Feedback Optimization for PID Controller Tuning

Traditional proportional-integral-derivative (PID) tuning methods often struggle in nonlinear systems with limited measurement availability. This study investigates PID controller tuning within a Riccati-based static output-feedback (SOF) optimization framework. Two SOF-based approaches—a gradient-based method and a mo...

Hakan Kızmaz · 0 citations
Open access Oct 2026

Adaptive MIMO fuzzy PID controller based on Runge-Kutta model

Nonlinear multi-input multi-output (MIMO) systems are difficult to control because of input–output coupling, measurement noise, and parameter variations. This study proposes a Runge–Kutta (RK) model-based predictive adaptive MIMO fuzzy proportional–integral–derivative (PID) controller for nonlinear dynamical systems. A...

İsmail Koray Gündoğan, Kemal Uçak · 0 citations
Open access Sep 2026

Pure Feedback Control for Nonlinear Systems Based on Predefined-Time Performance Function

This paper investigates the predefined-time adaptive neural tracking control problem for a class of nonlinear pure feedback systems with full state constraints, and proves that all closed loop signals are predefined-time-bounded and the tracking error meets the preset performance specifications.

Yang Li, Ya-Qi Yu, Quan-Min Zhu et al. · 0 citations
Open access 2026

Adaptive Neural Network Control of Fractional-Order Nonlinear Systems via Integer-Order Learning: A Memory-Efficient Approach

This letter develops an integer-order (IO) neural-learning approach for a class of fractional-order (FO) nonlinear systems. An initialization-compensated (IC) filtered tracking error yields an exact IO control-relevant error dynamics, while retaining the FO plant dynamics. This enables IO neural-weight adaptation witho...

Oussama Bey, M. Chemachema, T. Ahmed-Ali et al. · 0 citations
Open access Oct 2026

A Real-Time Hybrid Machine Learning–Based Adaptive PID Controller for Fixed-Wing UAV Lateral Stabilization with a Unified Validation Framework

Traditional Proportional–Integral–Derivative (PID) controllers are the dominant solution for Unmanned Aerial Vehicle (UAV) attitude stabilization due to their simplicity and low computational cost. However, their fixed tuning parameters limit performance under nonlinear dynamics, rapidly varying flight conditions, and...

I. Laabousse, Y. El Bouzekri El Idrissi, Mohamed Haidoury et al. · 0 citations
Sep 2026

Fixed-time adaptive sliding mode control for nonlinear systems based on rapid Kolmogorov-Arnold network

This paper proposes a fixed-time adaptive sliding mode control method for second-order uncertain nonlinear systems. The proposed controller combines fixed-time nonsingular terminal sliding mode control (FTNSTSMC) with a rapid Kolmogorov-Arnold network (RKAN). Unlike finite-time control, the fixed-time control design pr...

Shi-Xi Hou, Zi-Ru Xu, Yundi Chu et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.