Collaborative robots (COBOTs) can efficiently assist humans in interactions with robots and the environment when carrying out different tasks. They take advantage of the flexibility and cognitive decision-making skills of humans along with the speed, accuracy, strength, and reliability of robots. Interest in this research area has grown rapidly in recent years and there are applications in many areas, such as smart factories, health services, agriculture, service sector and surveillance. This paper reviews the state-of-the-art of planning and control strategies for collaborative robots. The survey includes various advanced approaches for motion planning, task planning, cooperative bimanual manipulation, cooperative mobile manipulation, learning from demonstration, exoskeletons and conjoined collaboration, and collaborative aerial robotics. A discussion of the challenges and future directions for the proposed research area are presented. This paper offers a comprehensive survey and insight for new researchers in the area.

In the domain of haptics, effective interaction with unknown environments is paramount. A key aspect of this lies in the pre-estimation of environmental impedance, a challenge addressed in this research study. Leveraging the paradigm of image-based object detection, this study introduces a novel methodology that correlates environmental impedance with object labels. By systematically applying this approach across diverse objects, the environmental impedance of unfamiliar entities can be reliably estimated based on shared image characteristics. This innovative technique not only advances the understanding of environmental interactions but also presents a proactive means of environment recognition. The methodology which is rooted in the fusion of haptics and computer vision, signifies a significant stride toward comprehensive and anticipatory environmental perception.

Autonomous manipulation in robot arms is a complex and evolving field of study in robotics. This paper introduces an innovative approach to this challenge by focusing on imitation learning (IL). In contrast to traditional imitation methods, our approach uses IL based on bilateral control, allowing for more precise and adaptable robot movements. Conventional IL based on bilateral control method relies on Long Short-Term Memory (LSTM) networks. In this paper, we present the IL for robots using position and torque information based on Bilateral control with Transformer (ILBiT). This proposed method employs the Transformer model, known for its robust performance in handling diverse datasets and its capability to overcome LSTM, especially in tasks requiring detailed force adjustments. A highlighting feature of ILBiT is its high-frequency operation at 100Hz, which significantly improves the system's adaptability and response to varying environments and objects with different hardness levels. The effectiveness of the ILBiT method is demonstrated through comprehensive real-world experiments.

Disturbance observers (DOb) are widely used to estimate disturbances, enabling robust position control through additional compensation currents and environmental force estimation. However, non-linear friction and gear compliance must to be excluded from the estimated disturbances torques for effective reaction force observation. Previous research has proposed an environment quarrier to estimate these torques by using an additional robot identical to the target. However, replicating the same robot is challenging due to cost and production variability. In this research, we used a gated recurrent unit to construct a virtual environment quarrier. This virtual environment quarrier was trained on free-motion data and applied to a reaction torque observer. We implemented the proposed method a 6-DoF manipulator and successfully regulated reaction torque.

Due to a declining population, Robots are expected to become reliable partners for humans in the near future. Such robots need to have the ability to manipulate objects dexterously in their hands. This paper proposes an in-hand manipulation method using modal control for realizing advanced tasks with tendon-driven robots. Using interaction mode control, position and force control can be switched in a unified manner. Therefore, it is possible to intuitively design a control system for when robots have contact with an object and when they do not. The proposed method uses a polar coordinate transformation to simply describe the translation, rotation, and grasping. Some experiments showed that not only a simple task such as the rotation of three objects with different stiffnesses, masses, and shapes, but also advanced tasks such as opening a cap can be realized using the proposed method. As a result, we demonstrated that the interaction mode control can be handled in the polar coordinate system and that the interaction mode control is effective for realizing dexterous object manipulation.

Recently, the development of electric vertical take-off and landing (eVTOL) has attracted much attention owing to their potential applications in various scenarios of human society, such as advanced transportation, surveillance, and inspections in hazardous environments. Among various eVTOL designs, the tilt-wing configuration stands out for its advantages, including reduced wing drag during ascent and effective utilization of propeller slipstream during transition phases. In this study, we propose a novel velocity control system designed in the wing coordinate system, employing an acceleration-based disturbance observer for tilt-wing eVTOLs. This method achieves enhanced disturbance rejection and decoupling control. The proposed approach was validated through computer simulations and hardware-in-the-loop simulations conducted in a wind tunnel.

Automation of loading explosives in mountain tunnel excavation is being promoted to solve the shortage of skilled technicians and the high incidence of industrial accidents. Loading explosives requires human-specific skills such as appropriate force, sensation, and adaptation to environmental changes, which have not been considered in previous studies. In this study, force-aware automation was realized that can adapt to environmental changes based on motion reproduction. Motion reproduction uses bilateral control to record and reproduce movements, and has the advantage of reproducing forces and teaching human skills to manipulators. Meanwhile, if the 3D relative position of a loading hole and a 6-DOF manipulator or the orientation or the unevenness of the loading hole changes from the recording mode, the recorded task will not be successful. Therefore, a method is proposed herein for detecting the 3D relative position, orientation, and unevenness of a loading hole, besides a method of motion reproduction with compensation for these changes to realize an automatic explosives loading system. The orientation is detected from the area of the ellipse extracted via image processing. Experiments confirmed that changes from the recording mode could be compensated for during reproducing so that recorded insertion motion could be successfully reproduced, even if the position and orientation changed.

To monitor processes of a tire changer that demounts a tire from a wheel, this paper proposes an end-effector with internally integrated force sensors and a process monitoring algorithm using its output. In the tire demounting processes, the tip of the proposed end-effector interacts with the tire and wheel rim. To measure the interaction force efficiently, the proposed end-effector is designed with a T-shaped beam for stress concentration, and the resulting strains along three orthogonal axes are detected by strain gauges. The sensor outputs are amplified and calibrated to decrease crosstalk and to improve accuracy. As a result, the proposed end-effector achieves high accuracy with a linearity of less than 1% FS. To utilize the proposed end-effector for process monitoring, an algorithm is developed and experimentally validated on a tire changer. The experimental results show that the proposed end-effector can correctly monitor and detect errors in the tire demounting processes using the proposed algorithm with a high success rate of 97%.

Fast and high-precision positioning is key in various industrial fields. In industrial machines with positioning mechanisms, drive systems are generally installed on the machine stands which are installed on the floor with support mechanisms. The machine-stand vibrations caused by the drive systems' high acceleration/deceleration motion deteriorate the positioning performance. Frequency-shaped final-state control (FFSC) is a method to suppress machine-stand vibration. In FFSC, the frequencies to be suppressed must be determined in advance. However, machine-stand vibrations have softening spring characteristics owing to the strong influence of the contact factors. It is unclear which amplitude is utilized when frequency characteristics are acquired in systems with softening spring characteristics because of their amplitude dependence. Here, a non-linear spring model that reproduces the softening spring characteristics is developed, and the frequency characteristics and positioning response are verified. Furthermore, the parameters of the FFSC in the non-linear model are optimized by the particle swarm optimization method, and the determination of the parameters is discussed.

This study provides an overview of Hitachi, Ltd.'s rescheduling and operation technologies focused on energy conservation, effective resource utilization for carbon neutrality and optimization and automation to address the decline in skilled workers. These developments are set against the backdrop of global carbon neutrality goals and the decreasing number of skilled workers due to a declining birthrate and an aging population. In the field of train traffic management, the “Dynamic Headway Solution” predicts demand by analyzing passenger flow and updates the train control timetable based on the results. Additionally, the “Hybrid traffic management AI” integrates the expertise of traffic controllers into rescheduling plans, ensuring the feasibility of the final output through machine-learning AI. These technologies enhance travel comfort and operational efficiency, compensating for the reduced number of skilled personnel. In driver assistance, “Technology for generating energy-saving speed profiles” and its application, “Technology to follow the speed profile as the application”, were developed to realize train operation, that adhere to energy-efficient speed profiles. Field tests of “Driver Advisory System” in TOKYO MONORAIL CO., LTD have demonstrated a total power consumption reduction of 16.4%. This technology supports global carbon neutrality goals.

This paper proposes a wireless power transfer (WPT) system utilizing flying capacitor converters (FCCs) on both the primary and secondary sides to reduce low-order frequency harmonic components of radiated emission. The FCCs produce a multi-level sinusoidal voltage with a few low-order frequency harmonic components. Consequently, the proposed system suppresses the harmonic components in the transmission coil currents that contribute to radiated emissions. Simulation and experiments with a 1-kW prototype demonstrate that the proposed system reduces odd-order harmonic components in the primary and secondary currents by over 10dB compared to conventional systems. Additionally, the proposed system reduces radiated emission by 22dB and 18dB at the third and fifth harmonics, respectively.

This paper presents a comprehensive analysis of zero-voltage-switching (ZVS) in the operation of a proposed peak current optimization scheme in dual-active-bridge converters with the voltage ratio between the primary and secondary sides is not matched with the turn ratio of its transformer. A minimal switching current and dead-time range are introduced to discuss the required conditions for ZVS realization in all possible operations with considering the parasitic capacitance of the power semiconductor device and dead-time. For a specific power range that can not achieve ZVS, a model-based lowest switching current control is proposed to achieve ZVS with ample switching current. Moreover, the allowable range of the dead-time for ZVS as function of the lowest switching current is also analyzed to determine the appropriate dead-time. Experiments using a 1-kW small-scale prototype were conducted to verify the theoretical analyses. We propose the use of silicon superjunction (SJ) MOSFETs for DAB converters by ensuring ZVS. To verify the concept, demonstrations using a fabricated DAB converter prototype with SJ-MOSFETs were conducted. It was verified that the proposed strategy can realize ZVS in a wide operating range, and achieve safe operation with SJ-MOSFETs.

Technologies for the optimization of train timetables have been investigated to reduce energy consumption. Previous studies have shown that the total energy consumption of all interstations is minimized when the target running time is selected at points that result in an identical gradient of the energy consumption against the running time. However, in certain cases, the running time cannot be selected at these points. In such cases, the conventional method cannot be used owing to the eccentric line profile, such as a steeper grade or longer distance intersection. In this study, we propose a method for setting the target running time at the lower limit of the interstation with small ∂E/∂T (small reduction of energy consumption per increase in target running time), and set the remaining target running times to other interstations to ensure that the gradient of all other interstations is identical. By verifying the proposed method through simulations, we confirmed that the proposed method can minimize the energy consumption by calculating all expected values of the gradient for the interstations. Therefore, our results shed light on dealing with line profiles for which a running time cannot be selected at the point that results in an identical gradient of energy consumption against the running time.

A new position sensorless vector control method allowing the SynRM to operate at maximum efficiency regardless of load and speed is presented. This method entails improving both position estimation accuracy and torque control accuracy by considering iron loss and inductance variations under a stator-flux-linkage-estimation-based loss minimization position sensorless vector control. The validity of this method is verified by experiments on a test machine. In addition, a method to compensate position estimation error caused by inductance variation due to cross-magnetic saturation is investigated. On the basis of some experiments, it is claimed that employing an approximation inductance function that accurately represents the actual inductance characteristics results in obtaining a high compensation effect for position estimation errors, regardless of whether the reciprocity characteristics of mutual inductance are taken into account or not.

A parallel resonant permanent-magnet synchronous generator (PMSG) system, which consists of a diesel engine, PMSG, full-bridge rectifier, and resonant parallel capacitors, has been proposed for series hybrid vehicles to save energy and reduce costs. In series hybrid vehicles, the parallel resonant PMSG system cannot adjust the output power; thus, charge-discharge control based solely on whether the engine is turned on or off is required. A charge-discharge control method using a state-of-charge map has been proposed that focuses on reducing the number of engine-starts. However, this conventional method is limited in its ability to extend battery life and improve fuel consumption in addition to reducing the number of engine-starts. Therefore, this study aims to further improve the charge-discharge control of parallel resonant PMSG system using reinforcement learning.

In this paper, we propose a new method to improve the drivability of tendon-driven systems. There are three main conventional tendon drive methods: (1) using one actuator per joint, (2) using springs for pre-tensioning, and (3) using two actuators to move one joint. The proposed method extends (2) and aims to avoid the traditional drawback of a two-inertia system. First, we conducted experiments on the load's tracking performance relative to the motor and demonstrated the effectiveness of using adjustments, including the spring-type mechanism. Subsequently, we evaluated drivability by examining the ratio between the input and received torque, confirming that the two-inertia effect was reduced compared to that in the conventional spring-type mechanism.

Due to the non-ideal environment in distribution networks with background voltage harmonics and various operation characteristics in the inverter system as control parameter variations, modulation dead zones, etc., the power quality of the grid-connected inverter in distribution networks can be severely affected by the above issues which results to total harmonic distortion (THD) and control oscillation issues. In order to provide power quality improvement and gird oscillation suppression, an improved active disturbance rejection control (ADRC) method is proposed in this paper. The proposed method first performs accurate estimation of the state variables using the Kalman filter and then provides robustness control of the inverter current with an improved differential feedforward ADRC algorithm. Theoretical analysis between the proposed method and conventional ADRC algorithm is also provided illustrating the benefit of the proposed method. The effectiveness of the proposed method is verified through mathematical analysis and the simulation results on a MATLAB/Simulink-based inverter platform. And the results have shown strong oscillation suppression and disturbance rejection capabilities with improved current THDs.

In this study, we proposed a Darlington power switch comprising a Si superjunction BJT (Si SJBJT) as a main transistor and an SiC-MOSFET as an auxiliary transistor (SiC-SJBMD), and the typical electrical characteristics of 7-A SiC-SJBMD of a breakdown voltage with 600V were experimentally investigated. The V_CE(sat) of the SiC-SJBMD is 0.84V at room temperature and 0.95V at 150°C at 7A. The power loss and delay time during the turn-off operation can be successfully reduced by decreasing the resistance in a charge-extraction circuit connected to the auxiliary gate of the SJBMD. The simulations revealed that the trade-off relationship between the V_CE(sat) and switching loss of the SiC-SJBMD is better than that of the SJBMD using a conventional Si-MOSFET as an auxiliary transistor. The simulations also compared the trade-off relationship of the SiC-SJBMD with that of the Si-IGBT and Si-SJMOSFET.

Increasing the switching frequency of SMPSs (Switched-Mode Power Supplies) can reduce their size and weight; However, developing high-frequency SMPSs with digital controllers is challenging. In conventional digital controllers, control outputs are updated every switching cycle, and therefore, if the switching period is shorter than the calculation time of the controller, the control cannot be achieved. Recently, multi-cycle control has been proposed to address this issue, where control outputs are periodically updated every multiple switching cycles. Despite its potential, the dynamic characteristics of multi-cycle controllers have not been thoroughly investigated. This study employs a multi-cycle voltage-mode PID controller in a trailing-edge modulated buck converter and investigates its small-signal transfer functions. The derived transfer functions are compared with the transfer characteristics obtained from circuit simulations and real circuit experiments in the form of Bode diagrams and time-domain waveforms, which validate the proposed model. The feedback gain is designed using the proposed model, and the results of the real circuit experiments indicate that this feedback gain achieves the design goal. The obtained results confirm that the proposed model is effective for controller design.

In recent years, synchronous reluctance motors (SynRMs) have been evaluated due to features such as being rare-earth-free, high-efficiency compared with other rare-earth-free motors e. g., induction motors, and maintenance-free. The field oriented control of AC motors has been generally used for instantaneous torque control. In SynRM, the dq-axes inductances vary significantly depending on the operating point, which results in a deterioration of the torque control performance based on field oriented control using the PI-gains designed by pole-zero cancellation without considering these variations. Here, the authors propose an innovative torque controller based on torque derivative value manipulation which has a low sensitivity against dq-axes inductance variance for high-performance torque control for SynRMs. The effectiveness of the proposed torque control system is verified through experiments and the sensitivity toward inductance variation is analyzed.

With aging pipe infrastructure, leakage management has become a critical issue for the sustainable operation of water supply. Several surrogate indicators and analysis methods are independently used to estimate leakage in a water distribution network. However, their accuracy leaves room for improvement due to their strict assumptions, such as linear leakage growth. This paper proposes a method for estimating leakage change by combining different indicators to achieve higher accuracy and consistency for the combined data. The proposed method is based on a state-space model of leakage change and its observation in Bayesian multilevel settings. A Markov Chain Monte Carlo (MCMC) sampler is used to estimate leakage and other parameters of the proposed model. The proposed method was applied to a case, and it was confirmed that the proposed method reduced the level of error in leakage estimation by half compared with an existing method.

Double-rotor motors are a promising configuration for electric aircraft propulsion, aiming to minimize the overall weight of the powertrain by leveraging their high-power factor. This study proposes a double-rotor motor with an axially extended closed-slot structure to secure the stator and reduce harmonic rotor losses. Subsequently, a novel cooling structure, termed as direct coil cooling, is implemented in the proposed motor, and a prototype small-scale stator is fabricated to evaluate the enhancement in current density. Lastly, a partitioned rotor core with carbon fiber reinforced plastic (CFRP) retaining ring is applied for high-speed operation.

The electrification of power sources for various types of mobility is progressing. In particular, high power density is required for air mobility applications require high power density. To increase the output density of high-speed fan-drive motors, reducing the temperature rise of the motor is crucial. We investigated various techniques to improve the output density of the high-speed fan-drive motor using amorphous metal, which reduces iron loss during the high-frequency drive, and using an air-cooling structure with a high-speed fan. Thus, a maximum torque of 200mN·m was achieved at a current density of 40Arms/mm²; moreover, a power density of 13kW/kg was obtained. This report presents the findings of our investigation and the evaluation results of a prototype motor.

One way to reduce the size and increase the power density of a motor is to increase its rotational speed. However, to drive motors at high speeds, the operating frequency must be increased, which increases in harmonic losses, primarily iron loss and AC copper loss. Therefore, harmonic losses must be quantitatively evaluated with high accuracy at the design stage. In this study, the validity of the method for evaluating the AC copper loss during high-speed operation is examined by comparing the difference between analytical values and evaluation formulas based on the relationship between the harmonic components of the voltage and current and slot shape.

This paper proposes a current-harmonics reduction method for a partial switching power-factor correction (PFC) converter with a proportional-sinusoidal (PS) controller for current control and a phase-shift control for an input current to reduce man-hours for adjustment of the partial switching. The PS controller realizes a stable current control for a peak input voltage larger than a DC-link voltage. In addition, the phase-shift control reduces the current harmonics, depending on the relationship between the input voltage and DC-link voltage. The validity of the proposed method was confirmed by a prototype. In the experimental results, the DC-link voltage and the input current are controlled according to the command value when the peak input voltage is larger than a DC-link voltage by using the current PS control. Furthermore, the current harmonics of the input current are within the limit of the IEC 61000-3-2 standard when the phase-shift controller is applied.

This study aims to improve the efficiency of an induction motor. The efficiency was measured under a load test using the inverter drive to verify the loss reduction effect when the winding changeover was applied. The stray load loss of series connection with a 184V synchronous 3 pulse could be reduced by 78% compared to the parallel connection with a 92V asynchronous pulse width modulation. Moreover, it was confirmed that the efficiency of the series connection is improved by approximately 4.4% compared to the parallel connection. Based on the above, it is confirmed that switching between series and parallel connections using a winding switcher can increase the efficiency of induction motors.