Next-generation high-precision mechatronic systems require safe and precise control of unmeasurable states. State-tracking iterative learning control (ILC) can achieve extremely high state-tracking performance up to the performance of state estimation, with convergence guaranteed apriori through the frequency-domain characteristics of the state estimator. The aim of this study is to develop a noncausal state estimation framework with verifiable frequency-domain characteristics. In batch-operated systems such as ILC, the use of noncausal design leads to substantial performance improvements that surpass the fundamental limits of causal approaches. Furthermore, by analytically verifying the frequency-domain characteristics of the noncausal state estimator, the developed framework retains the benefit of guaranteeing convergence in ILC. The developed framework is validated both by simulation and experiment, confirming improved state-tracking with monotonic convergence of ILC, achieved by exploiting noncausality in state estimation.

Magnetically levitated synchronous planar motors have gained widespread use in production environments owing to their fully non-contact multi-degree-of-freedom operation. However, achieving precise and stable motion remains challenging because of intrinsic instability, tracking delay, and cross-axis coupling. This study presents an integrated control system for a magnetically repulsive planar motor that combines feedback, feedforward, and decoupling controllers. The control design is based on system identification and modeling of the dynamics. In particular, the proposed feedforward and decoupling controllers are designed based on the identified motor dynamics, and explicitly account for the misalignment among the actuation point, the center of gravity, and the measurement point. Experiments using a one-dimensional planar motor validate the proposed integrated controllers.

Lithography systems for semiconductor and display manufacturing increasingly require high-speed, high-precision positioning. Based on a scanning-stage architecture driven by iron-core linear motors, controlled magnetic attraction is leveraged to provide lateral actuation without additional actuators. Because the same attraction introduces state-dependent thrust ripple, a feedforward framework is developed that couples iterative learning control (ILC) with Gaussian process regression (GPR) to generalize ILC-refined inputs to previously unseen trajectories. To reduce the burden of time-consuming ILC experiments and the rapid growth of GPR computation with training-set size, a trajectory optimization method is proposed that trades off GP predictive variance against computational cost, enabling accurate prediction with a compact training set. The proposed method is validated on a three-degree-of-freedom experimental platform, in which scanning is performed along the primary axis while lateral and yaw offsets are held constant. Experiments consistently reduce tracking errors across a range of scan conditions, supporting the proposed approach for high-performance lithography stages.

Condition-based maintenance (CBM) is an effective approach for ensuring the safe operation of power electronic converters used in industrial applications, grid-tied ones, and transportation systems. This paper proposes a condition-monitoring method of a DC-link capacitor used in a propulsion inverter for a rolling stock, where capacitance is used as the indicator of its degradation. This method utilizes the instantaneous output power of the inverter to reproduce the ripple current of the dc-link capacitor, which allows current-sensor-less condition monitoring of the dc-link capacitor. In addition, the capacitor voltage is obtained from a sampled one synchronized to a multiple of the carrier frequency available in a practical propulsion inverter. A 200-V 1.5-kW laboratory system with an induction motor is designed and constructed for emulating the propulsion inverter with synchronous pulse-width modulation in a region with constant output power. Experimental results confirmed that the capacitance of the dc-link capacitor was monitored from the sampled voltage and reproduced ripple current with the proposed method.

Pneumatic servo valves are essential for regulating flow rates in a wide range of industrial applications. This paper presents a parameterized feedforward control strategy designed to actively suppress pipe resonance, enabling high-speed and high-precision gas flow control. In the developed approach, feedforward parameters representing valve body dynamics and pipe resonance are identified using basis function iterative learning control (BF-ILC) and frequency-domain ILC (FD-ILC), respectively. This physics-based, stepwise learning scheme allows independent identification of the two components, minimizing parameter interference during BF-ILC. Experimental results demonstrate that the method reduces the settling time to 7 ms, over 80% faster than conventional feedforward control based on the inverse model. Furthermore, unlike FD-ILC, the proposed method maintains superior control performance across different flow rate trajectories while offering high interpretability. These results confirm that the method enables faster, more accurate, and task-flexible flow rate control in pneumatic valves.

Using multiple transmitters is an attractive option to supply power to receivers scattered over a large area in wireless power transfer systems based on resonant inductive coupling. However, multiple transmitters often suffer from changes in the amplitude and phase of the transmitter currents due to the influence of other transmitters and receivers, a phenomenon known as cross-interference. The change in the amplitude of the transmitter current can decrease the output power of each receiver and generate strong magnetic fields that do not comply with regulatory standards. In addition, the change in the phase of the transmitter current may reduce the power factor or cause hard switching of an inverter in the transmitter. This paper proposes novel multiple transmitters to solve the problems caused by cross-interference. The input voltage of each transmitter is controlled using a DC-DC converter to maintain the amplitude of the transmitter current at a constant value. Furthermore, a simple switching circuit is installed in each transmitter to automatically cancel out unwanted electromotive forces and variations in the self-inductance of the transmitter coil that affect the phase of transmitter currents. The results of an experimental evaluation validated the effectiveness and appropriateness of the proposed method by demonstrating that the system was able to transfer power without cross-interference.

Dielectric elastomers (DEs) are flexible capacitors consisting of a highly deformable dielectric membrane coated with compliant electrodes. DEs can function as actuators converting applied voltage signals into motion, sensors detecting changes in displacement or force based on capacitance variations, or even generators converting external mechanical stimuli into electrical energy. Actuation and sensing can also be performed simultaneously, achieving the self-sensing mode; this enables the implementation of closed-loop control architectures without introducing additional position or force sensors. The unique combination of DE characteristics, which include large deformation, high compliance, high energy density, low energy consumption, low cost, and self-sensing, enables the development of mechatronic actuators and soft robots that surpass conventional technologies in performance. However, the practical use of DE transducers in precise mechatronic and robotic applications is hindered by the strong input-output nonlinearity, which significantly complicates design, modeling, control, and self-sensing. This paper presents an overview of recent advances in model-based approaches and smart algorithms for DE systems. It discusses aspects related to system modeling, design optimization, motion/interaction control, position/force self-sensing, and sensorless control, focusing on engineering applications in the fields of micropositioning actuators and soft robotics.

We developed a method to compute emergency train timetables that minimize the total energy consumption. This approach supports the movement of trains during emergencies to adjacent stations using stationary energy storage systems during power failures. Determining the order for initiating emergency operations with minimum energy consumption is difficult because the potential number of combinations increases in proportion to the factorial of the number of trains. Therefore, we developed a computation method for determining the order of initiating emergency operations based on simple equations. An evaluation of the proposed method confirmed that it obtained a result that matched the optimum solution obtained achieved using the brute-force method in shorter time than brute-force method.

Geothermal power generation is a method of generating electricity using steam from underground hot springs. Compounds in this hot spring water, such as silicon dioxide and sulfur, accumulate on the inner walls of steel pipes at geothermal power plants, forming a layer referred to as hot spring scale. This deposition can clog the pipes in geothermal power plants. Therefore, determining the thickness of the hot spring scale deposited on the inside of steel pipes from the outside of the pipe is crucial. In this study, an inspection method is proposed to measure the thickness of hot spring scale inside a steel pipe using vibration induced by electromagnetic force. The proposed method was applied to steel pipes used in an actual geothermal power plant, demonstrating the usefulness of the approach.

This paper presents a performance comparison between consequent-pole and multi-monopole surface permanent magnet (SPM) bearingless motors. These topologies do not require rotor rotational angular positions for the magnetic suspension. The force error angles must be small; otherwise, the levitated rotor would be touched down on the stator. In fact, it is difficult to design the magnetic circuit to reduce the force error angle in the consequent-pole bearingless motor because the air-gap flux density includes modulated fluxes due to a salient-pole rotor core. In this paper, it is verified that the proposed multi-monopole SPM bearingless motor has great performances of high radial suspension force and low force error angle compared to that of the consequent-pole rotor. In addition, the design optimization is presented with respect to pole/slot combinations and several geometrical parameters of the rotor and stator.

Electrically- and hybrid-excited synchronous motors (EESMs/HESMs) are increasingly being favored over permanent magnet alternating current (PMAC) motors for high-performance applications due to their superior field weakening capacity, enhanced overload capacity, and reduced reliance on permanent magnets (PMs). However,EESMs/HESMs require controlled electrical excitation for their field winding in the rotor. Traditional brush and slip ring-based excitation methods are inefficient, require frequent maintenance, and suffer from reliability issues, including wear and tear. Although existingbrushless excitation systems (BESs) incorporate auxiliary machines or modify motor designs to eliminate brushes and slip rings, most rely on controlled switches placed in the rotor to reverse the polarity of the field current, necessitating complex wireless communication between the stator and rotor. To address these challenges, this study introduces a novel BES forEESMs/HESMs. The proposed system can control bipolar field current (BFC) entirely from the stator without any rotor-side controllers or wireless communication. First, the operating principle of the proposedBES, which comprises a stationary-side series resonant converter (SRC) in conjunction with a rotary transformer (RT) having stator-side primary, and two rotor-side secondary windings, is explained. Secondly, a selective multiple excitation frequency control method is presented, which utilizes selectively tuned resonant tanks in each of the secondary windings to achieve bipolar field current control entirely from the stator. This approach obviates the need for any communication between the stator and the rotor. Finally, the performance of the proposed system for a 105kW, 6000rpm segmented-rotor HESM is evaluated using a co-simulation framework in which the RT and HESM are modeled in ANSYS Maxwell, whereas the BFC converter is modeled in the MATLAB/Simulink and ANSYS Twin Builder. The proposed method also helps enhance the torque density of HESMs.

Solar photovoltaic (PV) technology is a promising renewable energy source; however, its efficiency considerably reduces dust accumulation, particularly in arid or polluted environments. Traditional water-based or manual cleaning methods are inefficient, costly, and unsustainable. This study proposes an automated dry-cleaning system that integrates machine learning and image processing for real-time dust detection and targeted cleaning. The system employs Canny edge detection and YOLOv5 to isolate PV panel regions and classify dust severity, labeling panels with ≥30% surface coverage as “dusty.” Arduino Mega and Jetson Nano jointly control a brush-based mechanism driven by dual DC gear motors, optimized for vertical (400rpm) and horizontal (200rpm) cleaning. Experimental evaluation on 15-W PV modules demonstrated that the system improved module efficiency from 77.1% under dusty conditions to 96.0% after cleaning, corresponding to an average 39% increase in output power. The dust detection algorithm achieved 82% classification accuracy, and the optimal cleaning performance was obtained with two sweeps (≈1min 59s), beyond which additional passes yielded negligible gains. By integrating accurate detection, optimized automation, and sustainable dry-cleaning, the proposed framework provides a scalable theresource-efficient solution for maintaining long-term PV performance.

Setting the stiffness parameters of a control system in contact-rich tasks is a critical factor when designing physical interaction/contacts with its environment. Therefore, studies on active compliance are being actively conducted across a wide range of domains, including industrial applications. The asymmetric design of stiffness parameters in active compliance can potentially enable interactions that are not possible with conventional methods of either active or passive compliance. However, asymmetric stiffness does not satisfy passivity, stability of the system is the issue. This study focuses on the stability of compliance control in a robot arm with an asymmetric stiffness matrix. The convergence stability of the admittance control is discussed. The paper presents the derivation of an asymmetric stiffness matrix and its incorporation into the admittance model. The experimental results validate the effectiveness of the proposed method and realize stable control using asymmetric stiffness.

In biomedical engineering and microelectronics, microgrippers have been widely used for micromanipulation. Various reports have been made on the gripping of objects using a piezo-driven microgripper. In these studies, the gripping operation was performed by switching the controller before and after contact with the object. However, the force transmission was not achieved using these conventional methods. Therefore, these conventional methods may result in damage to the object. For the realization of safety micromanipulation, perceiving the force feedback of the microgripper is one of the important technologies. Therefore, this study proposed a micro-macro bilateral control using a piezo-driven microgripper. For the bilateral control in the microgripper, the position and contact force were estimated using strain gauges. Additionally, this study developed the hysteresis compensator based on an observer theory. The hysteresis characteristics caused by the piezoelectric actuator were compensated as part of the disturbance. The usefulness of the bilateral control in the microgripper was confirmed from the experimental results.

This paper explores the control of a motorcycle using automatic tuning based on Bayesian optimization. Motorcycles exhibit high stability at high speeds owing to the gyroscopic effect. However, this effect diminishes at low speeds, forcing riders to rely on weight shifting or steering inputs to maintain balance. To improve safety in low-speed scenarios, autonomous regulation of motorcycle stability becomes essential. Existing self-balancing methods include applying a linear quadratic regulator (LQR). This approach involves creating a specialized low-speed region model using SPACAR, a finite element method-based program, and applying LQR control to achieve stabilization. However, when this approach is implemented on an actual motorcycle, unmodeled dynamics, such as tire friction, can hinder control. Therefore, gains must be fine-tuned through trial and error on actual motorcycles. This paper presents a method using Bayesian optimization for adjusting controller gains more efficiently. Through simulations and experiments, we demonstrate that Bayesian optimization can explore a wider range of parameters and find better parameters than manual tunning.

Task planning for robots often involves defining a wide range of possible actions, which makes the process both time-consuming and labor-intensive. This study presents an automated task-planning framework for teleoperated grasping robots, leveraging large language models (LLMs). Task instructions are provided via voice input, while object detection is performed using the No-Label Detection System (NLDS), which integrates YOLOv8 for coordinate detection and GPT-4o for semantic labeling. This configuration allows the system to flexibly recognize previously unseen objects and align visual outputs with natural language commands. The proposed framework comprises three main stages: (1) task planning based on operator instructions, (2) object detection and extraction, and (3) grasp position estimation. Experiments conducted on a physical robotic system demonstrate the framework's capability to interpret ambiguous commands and manage overlapping objects, achieving robust performance in complex, real-world scenarios.

Although cell manipulation technologies have been adopted in a wide variety of applications, at present operators must rely solely on visual information to perform such procedures without damaging the cells. Consequently, there is an increasing demand for the integration of tactile feedback. However, conventional force sensors face significant challenges in detecting ultrafine forces exerted on soft targets such as cells. In this study, we propose an admittance-based bilateral control with reaction force estimation using an optical tweezer. The results of an experimental evaluation showed that a prototype system was able to estimate and transmit forces on the order of piconewtons to facilitate ultrafine teleoperation. Thus, our results show that the proposed approach overcomes conventional limitations in measuring ultrafine forces and enables intuitive force feedback. These capabilities are expected to reduce the difficulty of specialized microscale tasks such as cell manipulation significantly.

In factories that manufacture round bars, an accurate count of the number of bars is required before shipment. Manual counting is time-consuming and labor-intensive, and the larger the number of bars, the less accurate is the count. Therefore, we automated the counting process using semantic segmentation. The method introduces a novel idea to reduce the effort of annotating the training images and generating the ground truth. Specifically, points are manually annotated on each end face of a round bar, and ground truth is automatically generated from the points. The segmentation model was trained using the generated ground truth to extract each end face from the target image. The round bars were counted by applying labeling after removing salt-and-pepper noise. To confirm the effectiveness of our method, an image dataset was created. In experiments, the performance of DeepLabV3+ and Unet++ with conflicting features were compared. The results showed that Unet++, which considers local information, performed better.

Active gate drivers (AGDs) enhance the trade-off between switching losses and voltage surges by actively controlling switching transitions. However, conventional AGDs, that vary gate resistance to adjust switching behavior, are limited in their capability to operate at high speed. This paper proposes a novel AGD architecture that combines a current source for the turn-on transition and an active gate resistor for the turn-off phase. The current source enables faster switching compared to resistance-based control and the active gate resistor employed during the turn-off phase is similar to conventional AGDs. The results of double-pulse testing show that the proposed AGD reduced turn-on losses by up to 32.4% compared to traditional gate drivers and by 6.55% compared to existing AGDs, while maintaining similar turn-off losses.

This paper proposes a single- and three-phase compatible AC-DC converter with a simple distributed control scheme. The proposed system consists of independently controlled single-phase AC-DC converter modules, connected in series and/or parallel. This modular architecture enables flexible configurations with scalable capacity for both single-phase and three-phase operation, without requiring high-speed communication. In conventional series-connected systems, interactions among current controllers can cause overmodulation. To address this issue, the proposed control strategy extends droop control to the current control loop, enabling uniform load sharing even in series-connected configurations. A common control strategy is employed for both single-phase and three-phase operation, eliminating the need for control system modifications. The proposed architecture was experimentally validated using six 1.2-kW AC-DC converter modules. A 7.2-kW three-phase AC-DC converter, configured with two modules in series on the AC side and six modules in parallel on the DC side, operated successfully under rated conditions without overmodulation. Moreover, the distributed control scheme maintained continuous operation without any additional control actions even in the event of a failure in a series-connected module.

This paper proposes an auxiliary resonant commutated pole (ARCP) converter for wireless power transfer (WPT) systems to suppress low-order harmonics in radiated emissions. To further reduce low-order harmonics, selective harmonic elimination pulse width modulation (SHE-PWM) is applied to the ARCP converter. The switching angles for SHE-PWM are determined through numerical analysis, and the switching patterns for 3- and 7-pulse PWM are derived. The effectiveness of the proposed system in reducing radiated emissions is validated using a mini model. The results show that the ARCP converter combined with 3-pulse PWM successfully suppresses third- and fifth-order current harmonics and achieves zero-voltage switching (ZVS) for all switching events. Furthermore, radiated emission measurements confirm reductions of 21 and 24dB in the third and fifth harmonic components, respectively.

In this study, a novel rotor structure for V-shaped interior permanent magnet (IPM) motors was proposed. V-shaped magnet arrangement enables the generation of concentrated magnetic flux on the d-axis, resulting in a high torque density. However, reducing the leakage flux generated in the outer bridge was an important design issue. In the proposed motor, trapezoidal magnets were inserted into the rotor core instead of conventional rectangular ones. By gradually narrowing the width of the magnets in the radial direction, the magnets can be supported by the rotor core without outer bridges. In this study, numerical verification was performed to verify its effectiveness. The numerical verification confirmed that the volume of magnets required to obtain the desired torque was reduced by 10.2% compared with that of conventional V-shaped IPM motors. In addition, a prototype was used to demonstrate the effectiveness of numerical verification. It was anticipated that the positional variation of the magnets in the rotor core would be a problem with this structure; however, during prototype production, it was demonstrated that this issue could be resolved by using a jig to fix the magnet position.

Capacitive power transfer (CPT) systems have recently attracted considerable attention as a means of supplying electric power to rotating bodies. Conventional CPT systems typically incorporate separate rotary capacitors and series-resonant inductors, which increase structural complexity and system volume. In this study, a novel coaxial coil electrode (CCE) coupler is proposed to address these limitations. The CCE coupler consists of two coaxially arranged coil electrodes, utilizing both capacitive coupling between the inner and outer coils and their inherent self-inductance. By integrating the functions of a rotary capacitor and a compensation inductor into a single, unified structure, the CCE coupler enables a simplified and space-efficient system design. The capacitance and inductance were designed and analyzed using finite element method (FEM)-based electromagnetic field simulations, and a prototype was fabricated to validate the design. The measured inductance and capacitance were well explained by the simulation results. Furthermore, the power transfer performance of a CPT system incorporating the CCE coupler was evaluated through both circuit simulations and experimental testing. The measured output power and transfer efficiency exhibited trends consistent with simulation predictions, confirming the feasibility and effectiveness of the proposed CCE coupler for powering rotating bodies.

In today's business environment, companies must balance cost efficiency with customer satisfaction through effective supply chain network design. The Facility Location Problem (FLP) addresses this challenge, but becomes computationally intractable for large-scale instances involving multiple product categories and heterogeneous warehouse sizes. To address these limitations, we propose a mathematical model and an efficient solution approach that substantially reduces computational time while preserving optimal facility location decisions. The proposed method integrates relaxation techniques to improve computational efficiency, jointly determining warehouse locations, capacities, and product assignments under storage constraints. Unlike classical LP-based relaxation methods and LP-guided variable selection approaches used in capacitated and step-cost FLPs, our approach exploits the structural properties of warehouse-product interactions, enabling tractable solutions without compromising modeling fidelity. A European case study demonstrates that the proposed approach significantly reduces computational time without loss of accuracy. This research provides a practical decision-support tool for enhancing supply chain resilience and efficiency.

In this paper, the authors proposed the circuit model for obliquely arranged cables. The cables were divided into short segments, which were parallel to the horizontal and vertical conductor planes as reference ground, and these segments were connected in series to construct the overall model. The model parameters of each segment were determined via electromagnetic field simulation using quasi-static approximation and data interpolation. To validate the proposed model, an obliquely arranged transmission line was measured using time domain reflectometry (TDR), and the results were compared with the circuit simulation results obtained using the finite difference time domain (FDTD) method. The results showed that the variation in line constants attributed to the oblique arrangement could be determined through TDR measurements. The line constants obtained using the circuit model agreed with the TDR measurement results, which confirmed that the proposed circuit model was appropriate. The frequency-dependent behavior of the circuit model was verified by comparing the measured reflection of the common mode with the calculated results using the circuit model operated in the frequency range of 0.15MHz-500MHz. The calculated results were similar to the measured results, confirming the validity of the proposed circuit model.

The tendency of metaheuristic algorithms to converge prematurely during surface-mounted permanent magnet synchronous motor (SPMSM) parameter identification compromises their accuracy. To address this issue, this paper proposes an improved eel and grouper optimization (IEGO) algorithm. Building on the standard EGO, IEGO implements three key enhancements. First, an elite dynamic reverse learning strategy during initialization elevates initial population quality and search efficiency. Second, the integration of the sparrow search algorithm surveillance mechanism with adaptive normal cloud modeling optimizes position-following strategies to improve global exploitation capability and convergence speed. Lastly, the incorporation of Tent chaotic mapping strengthens local optima avoidance to prevent premature convergence and boost global exploration. During parameter identification, the measured motor signals (current, voltage, and speed) are processed by the identification model, wherein the IEGO algorithm performs iterative optimization using a fitness function, enabling rapid, high-precision identification of critical parameters, including stator resistance, d/q-axis inductances, and permanent magnet flux linkage. Simulation and experimental results confirm the significant advantages of IEGO in SPMSM parameter identification, demonstrating superior accuracy, enhanced convergence stability, and reliable parameter foundations for precision motor control.

In this letter, a new capacitor boost converter for a switched reluctance motor is proposed. The proposed converter can boost the motor voltage without additional switching components for an asymmetric half-bridge converter. The basic operation of the proposed converter is verified by comparison with a previously proposed converter.