Active Disturbance Rejection Control in Nonlinear Dynamical Systems

Active disturbance rejection control (ADRC) is a model-independent paradigm that achieves robust regulation of nonlinear dynamical systems by estimating and compensating for uncertainties and exogenous disturbances in real time. At its core lies the extended state observer (ESO), which augments the conventional state vector to include a ‘generalised disturbance’ term comprising all unmodelled dynamics, external inputs and parameter variations. By cancelling this disturbance online through a feedback loop, ADRC avoids reliance on precise mathematical models and offers resilience against varying operating conditions. Recent advances have extended the basic linear ESO to nonlinear formulations (NLESO) that mitigate observer peaking and enhance convergence speed, and have integrated ADRC with input–output feedback linearisation theories to yield scalable control laws for systems with known relative degree. The global significance of ADRC stems from its broad applicability to fields from robotics and renewable energy systems to automotive and aerospace platforms, where rapid adaptation to changing loads and uncertainties is critical for performance and safety.

Research from Nature Portfolio

No recent Nature Portfolio content available.

Research from all publishers

Recent studies have advanced ADRC across several fronts. A 2019 investigation introduced a finite-time nonlinear extended state observer in an ADRC scheme for permanent magnet DC motors, demonstrating mitigated peaking phenomena and proven stability via finite-time Lyapunov analysis under friction and measurement noise. A 2020 model-free active input–output feedback linearisation approach applied improved ADRC to a flexible joint manipulator, embedding a disturbance-estimating observer within a feedback linearisation law to transform the nonlinear system into a chain of integrators using only relative degree information. Another 2020 contribution proposed a nested linear extended state observer within ADRC for uncertain single-input–single-output plants, placing an outer observer in parallel to address bandwidth constraints and achieve substantial reduction in tracking error without increasing noise sensitivity.

Active Disturbance Rejection Control in Nonlinear Dynamical Systems publication trend

The graph below shows the total number of articles in active disturbance rejection control in nonlinear dynamical systems across all publications each year (not limited to Nature Index journals).

Technical terms

Active disturbance rejection control (ADRC): A control strategy that estimates and compensates for internal and external disturbances without requiring an accurate system model.

Extended state observer (ESO): An observer that augments the system state to include disturbances and uncertainties for online estimation.

Nonlinear extended state observer (NLESO): An ESO employing nonlinear functions to enhance disturbance estimation, reduce observer peaking and improve robustness.

Generalised disturbance: The aggregate term comprising all unknown dynamics, external disturbances and modelling errors in the control loop.

Input–output feedback linearisation: A method that transforms a nonlinear system into a linear equivalent by cancelling nonlinearities through feedback and coordinate transformation.

References

1. Speed Control of Permanent Magnet DC Motor with Friction and Measurement Noise Using Novel Nonlinear Extended State Observer-Based Anti-Disturbance Control. Energies (2019).
2. Model-free active input–output feedback linearization of a single-link flexible joint manipulator: An improved active disturbance rejection control approach. Measurement and Control (2020).
3. Novel Active Disturbance Rejection Control Based on Nested Linear Extended State Observers. Applied Sciences (2020).