In ISA Transactions, researchers from the Changchun Institute of Optics, Fine Mechanics and Physics, Chinese Academy of Sciences, report a compound anti-disturbance method for servomotor position control. The study integrates a novel adaptive sliding-mode disturbance observer (NASMDO) with an adaptive non-singular terminal sliding-mode controller (ANTSMC) to keep tracking accuracy high when disturbances change rapidly. Across device-level trials, the approach consistently lowered tracking error compared with widely used disturbance observers and sliding-mode controllers, while maintaining finite-time stability guarantees important for engineering deployment.

The problem under study is familiar to precision platforms such as aviation optoelectronic turntables: wire winding, friction, carrier sway, and wind loads introduce multi-source, time-varying disturbances that degrade pointing and tracking. Existing frameworks—internal-model control (IMC), active disturbance rejection control (ADRC), and disturbance-observer-based control (DOB)—each help, but they face trade-offs in modeling effort, bandwidth tuning, and stability proofs under fast disturbances. Classic sliding-mode controllers add robustness but can chatter; terminal sliding-mode designs aim for fast convergence but may encounter singularities. The paper positions NASMDO+ANTSMC as a way to reconcile these tensions by pairing fast disturbance estimation with smooth, finite-time tracking.

The NASMDO was designed to estimate the total disturbance—external loads and model uncertainty—using velocity information, not only position. That choice let the observer construct a dynamic sliding manifold tailored to rapid changes. A sliding-mode-assisted term based on a nominal model then compensated the remaining estimation error, and the authors proved that the disturbance-estimation error converged into a bounded set in finite time. In practical terms, the observer produced a compensation signal that closely tracked time-varying disturbances without relying on fixed, excessively high gains that typically aggravate chattering.

Around this observer, the team built ANTSMC, a non-singular terminal sliding-mode controller with an adaptive approaching law. The law adjusted its switching gain online: it increased when the system faced larger transient errors and decreased near steady state, reducing the amplitude of the sign term that drives chatter. A second Lyapunov analysis established that the closed loop remained finite-time stable and that the adaptive gain converged toward small values once tracking settled. This pairing preserved the well-known robustness of sliding-mode control while taming its most common side effect in precision drives—chattering.

Validation took place on an ARM-based servomotor turntable equipped with a hysteresis brake to generate realistic, time-varying loads. The researchers compared NASMDO to a conventional DOB and to an adaptive sliding-mode-assisted observer (ASMADO). They also compared ANTSMC against two tracking baselines: a standard sliding-mode controller (SMC) and a non-singular terminal SMC (NTSMC). Test scenarios included sinusoidal and square-wave disturbances spanning 0.5–30 Hz, capturing both smooth drifts and abrupt torque steps. Time-series plots and summary tables show that the NASMDO-based system produced smaller tracking errors and calmer control inputs across conditions, and that the adaptive gains in both NASMDO and ANTSMC settled to low levels after transients—visual evidence that unnecessary switching activity was suppressed.

Quantitatively, the paper reports that the NASMDO combination reduced tracking error on average by more than one-third relative to commonly used observers under mixed disturbance types and frequencies. When the observer was fixed to NASMDO and only the tracking law changed, ANTSMC further reduced RMSE by more than 40% versus SMC and by more than 30% versus NTSMC, with repeated runs confirming consistency. While the article avoids overselling, the overall picture is clear: by incorporating velocity into the disturbance-observer design and by adapting switching gains, the compound architecture provides a realistic balance of robustness, speed, and actuator smoothness for servo motor anti-disturbance control.

From a broader perspective, the work is relevant wherever precise, predictable motion under uncertainty is required—line-of-sight stabilization, precision robotics, and laboratory instruments are immediate examples. The proofs for practical finite-time behavior offer timing predictability, while the modular design means the observer and controller can be combined with other elements if application constraints differ. Because the method was implemented and tested on an embedded platform with transparent parameter-tuning guidance, the path from paper to pilot deployment appears straightforward for teams already using DOB or sliding-mode designs.