Analysis and Design of Sensor-Driver-Aware Integral Nonsingular Terminal Sliding Mode Control for Buck Converter Power Interfaces in Actuator Systems

Buck converter power interfaces are commonly used as local voltage regulation units in compact actuator-driven microsystems, where the regulated voltage needs to remain stable under input, load, and circuit-level disturbances. In practical control loops, passive-parameter variations, external perturbations, and non-ideal sensor-driver dynamics may distort feedback signals, delay effective duty-cycle action, and degrade the transient response of conventional robust controllers. Motivated by this issue, this paper presents the analysis and design of a sensor-driver-aware integral nonsingular terminal sliding mode control (INTSMC) method for a buck converter power interface under multi-source disturbances. A control-oriented averaged model is first constructed by incorporating converter parameter perturbations, load-side disturbances, Hall sensor dynamics, and isolated driver characteristics into a unified representation. Based on this model, an integral nonsingular terminal sliding surface is designed to improve voltage tracking performance while avoiding singularity in the reaching process. The corresponding control law is further arranged in a pulse-width modulation-realizable duty-cycle form, making it suitable for digital converter control. In addition, a phase-trajectory-based response-time estimation method is introduced to analyze the influence of initial states, disturbance levels, and hardware dynamic parameters on the closed-loop reaching behavior. Simulation studies under different operating conditions are carried out to evaluate the proposed controller. Simulation and experimental results show that the proposed method achieves a settling time within 33 ms, a steady-state voltage error within 0.01 V, and a measured efficiency of 83.5%~88.9%, indicating its feasibility for robust power regulation in micro-actuator-oriented microsystems where sensor-driver dynamics cannot be ignored.