Induction motors are required to operate at a high fundamental-to-switching frequency ratio (fe/fs) for high-power drives. As fe/fsincreases, the current control system significantly deteriorates due to the inaccuracy of the conventional discrete-time model caused by discretization phase error. To address the issue, the phase-error-free discrete-time model (PEFD model) has been proposed. In implementation, the PEFD model requires polynomial approximation, yet the methodology for determining optimal approximation orders at different operating conditions remains uninvestigated. Hence, this study analyzes the relationship between the approximation accuracy and operating conditions. Then, the proper approximation orders for different operating conditions are recommended. Finally, simulations and experimental validations are conducted.

In this study, a new Extended EMF observer is proposed for a Dual Three-Phase Synchronous Motor, and position-sensorless control in all speed ranges is achieved with a single linear observer. By utilizing Vector Space Decomposition, two Extended EMF vectors are defined, which are excited by the rotor speed and high-frequency injected signal, respectively, and they are separated into two different coordinates. The voltage vector sum of them is redefined as ${\text{EEM}}{{\text{F}}_{S{D_{\omega + h}}}}$, which can be estimated by a single linear observer in all speed ranges. Compared with the conventional method for three-phase synchronous motors, the high-frequency injected signal in this method does not affect the fundamental signal, even during the middle-speed range, and torque ripple is avoided. In this study, the linear observer of ${\text{EEM}}{{\text{F}}_{S{D_{\omega + h}}}}$ was configured, and its performance was evaluated by comparing it with the conventional method for three-phase motors.

A multiphase induction motor (IM) can change the pole number by controlling it as a superposition of 3phase IMs with different pole numbers through vector space decomposition (VSD). However, during pole change, excessive current is generated, which can trigger the current limiter, causing torque fluctuations. In this paper, to address this, a new pole change method applying maximum torque per ampere (MTPA) control between different pole numbers, minimizing the current during pole change, is proposed. This method prevents current from reaching its limiter and expands the region where pole change can occur at a constant torque. The efficacy of the method is demonstrated through simulations using a 400kW IM.

This paper presents the Funabot-Suit for the Upper Body, enabling users to perceive seven upper body motions (shoulder abduction, elbow flexion/extension, and trunk rotation) through kinesthetic feedback induced by 80 McKibben artificial muscles embedded in elastic clothing. Unlike existing systems that rely on rigid structures or vibrotactile feedback, the Funabot-Suit provides soft, multi-joint kinesthetic feedback via garment deformation. Building on our previous work focused on trunk motion, this study extends perception to more localized joints such as the shoulders and elbows. The experimental results from nine participants indicate that larger body parts tend to induce stronger and clearer sensations, while certain localized motions, such as elbow extension, are perceived less clearly. These findings suggest that both the size of the body part and the placement of artificial muscles affect kinesthetic perception. The results provide design insights for future wearable devices targeting VR, rehabilitation, and human-robot interaction.

This paper presents a wearable haptic device that can perceive eight directions of the forearm utilizing cloth deformation. Inducing kinesthetic perception through wearable haptic devices is expected to have applications in VR, rehabilitation, and teleoperation. The demand for haptic devices made of flexible materials that can provide intuitive feedback has rapidly increased. In this paper, we developed a novel haptic device for the forearm that consists of thin artificial muscles and existing clothing. The cloth deformation caused by the contraction of the artificial muscle allows the wearer to perceive four directions of the forearm, and the results of the perception experiment showed 98.8% accuracy. Furthermore, experimental results showed that synthesis stimuli can extend the wearer’s kinesthetic perception in eight directions. The findings provide important guidelines for the design of future haptic devices.

This manuscript investigates the impact of unmanned aerial vehicles (UAVs) and their cameras’ key specifications on the efficiency of infrastructure visual inspections using UAVs. The challenges of UAV-based inspection data collection include reducing necessary waypoints to improve inspection efficiency while collecting desired data for sure. Existing studies generally concentrated on optimizing the efficiency of UAVs’ inspection trajectories and waypoints when equipment is given. However, the impacts of equipment specifications and their selections have not been investigated, yet they also significantly affect inspection efficiency. This manuscript surveys the key specifications of current equipment available for inspection and investigates the variation in efficiency through inspection simulations with different equipment specification combinations. Simulation results reveal that with the highest pose accuracy achievable by existing UAV platforms, it is possible to achieve efficient inspection. However, such high pose accuracy may be difficult to achieve, underscoring the importance of targeted selection of camera focal lengths according to UAV specifications and the types of pose errors. Additionally, using higher-resolution cameras may substantially improve the efficiency of high-resolution data collection tasks, which is a more practical solution for enhancing efficiency than further improvements in UAV pose accuracy. The conclusions drawn from this research provide actionable guidelines for equipment selection and development and lay the foundations of efficiency modeling, ultimately contributing to the advancement of UAV-based inspection methodologies.

Compliance is important for the rehabilitation of wearable assistive robots. Various methods have been studied to improve the compliance of wearable robots, such that the use of impedance control of actuators and soft actuators with high compliance. On the other hand, for robots whose degrees of freedom of movement are limited by rotational axes like conventional robots, compliance performance may deteriorate due to mounting misalignment even if the actuator is designed optimally. Soft wearable devices that do not have rigid joints and are composed of soft base materials also exist, but the force is dispersed and cannot exert great force. Our Funabot- Plate consists of a base cloth, McKibben-type pneumatic artificial muscles, and thin, rigid plates that efficiently transfer the exerted force of the artificial muscles to the human body. Thus, Funabot-Plate expects a high output efficiency while high compliance. To verify the structural validity of the Funabot- Plate, two experiments were conducted to quantitatively compare its compliance characteristics with the rigid-frame robot. Through the measurement experiments, we were able to quantitatively evaluate that the structure of Funabot- Plate exhibited higher compliance in both conditions, transparent mode and assist mode, when compared to conventional rigid-frame robots.

Fabric actuator (Funabot) is expected to be applied to the haptic device. On the other hand, conventional control method in fabric actuator only feeding back the surface shape by depth sensor, although haptic device evokes the tactile via the direct contact between the device and the human body. We developed the haptic device for grabbing and established the leader-follower system that include force distribution sensors and realized measuring the leader-side tactile as the distribution information, sending the follower-side and tracking the static distribution by feed-back control at the follower-side. The basic performance was verified by two experiments with the soft torso made of silicone in the static condition. The first experiment clarified that our device had an independent controllable tightening force of at least 7.9 N for seven channels in a 2 cm width. The second experiment proved that our device could track the leader tactile force within the error of 0.21 N in most cases. While the tracking performance was admitted, response performance was slow because of the characteristics of artificial muscle.

Assistive robots with soft actuators do not have joints like exoskeletons and have attracted attention because of their affinity for humans. In this study, we propose a soft assistive suit called "Funabot-Plate," which is driven by artificial muscles, has no joints, and is partially equipped with rigid plates and contact force distribution sensors. Funabot-Plate is expected to be applied to rehabilitation in the future, using contact force distribution information to evaluate and control suit safety. Assistive torque was designed based on the geometric model. Joint angle estimation and control were performed using contact force distribution sensors. The range of motion, contact force on the wearer, and convergence performance of the step response were evaluated through several basic experiments. As a result, it was confirmed that the contact force was measured normally, and the response converged with a slight error of about 1.6 deg. to the command value, although the range of motion remained a problem.

In Japan, where the population is aging, traffic accidents caused by elderly drivers have become a social problem. The main cause of the accident is a decline in cognitive ability, and there is an urgent need to develop technology for early detection of decline in the ability. The authors have been developing a technology to evaluate the cognitive ability of a driver from the driving behavior data. In the driving behavior data, there are some scenes where the cognitive ability can be evaluated and some scenes where the cognitive ability cannot be evaluated. Therefore, in this paper, we propose a clustering method for driving behavior data. Here, it is considered that the cognitive ability is the accuracy of the operation for the surrounding situation, and it is useful to cluster the relationship pattern between the situation and the operation for the driving behavior data as a group. In this method, Coupled-GP-HSMM and Convolutional Variational Auto Encoder are applied as a time series clustering method. This method realizes clustering of the relationship pattern between the time-series data representing the situation and the time-series data representing the operation.

This study presents current control algorithm based on a finite control set model predictive control (FCS-MPC) to achieve both high dynamics and current ripple suppression. In the proposed method, the smoothed voltage vectors with a finite set are applied as a control input candidate to avoid a sudden change in output voltage which generates large current ripple. In addition, the smoothness is determined automatically depending on a drive situation and system’s specification. Owing to this, fast transient response is achieved while keeping small current ripple during drive operation. The simulated and experimental results obtained with a permanent magnet synchronous motor (PMSM) show that the proposed method is effective for current ripple reduction and high dynamics control as compared to traditional FCS-MPC approach.

This paper proposes a capacitor current reduction method using real-time model predictive control (MPC) for a double three-phase permanent magnet synchronous motor (DTPPMSM) drive system. Capacitor current reduction for volume minimization of DTP-PMSM drive system is important issue. Optimizing switching timings by MPC is promising approach. However, it is difficult to implement ideal MPC in a real system because it requires a huge amount of calculation to predict the capacitor current. In this paper, we propose a new method that is possible to search optimal switching timing for reducing the capacitor current by using the MPC, whose calculation amount can be implemented in real-time. This paper shows the possibility of realizing an capacitor current reduction method with high control performance by reducing the amount of computation for MPC.

This paper proposes a torque ripple estimation and reduction of IPMSM with concentrated winding considering nonlinearity of magnetization characteristics. First, the proposed torque ripple estimation method which ignores nonlinearities such as magnetic saturation and cross-coupling is derived and some problems of this approach are presented. Then, the torque ripple estimation method considering nonlinearity of magnetization characteristics is proposed, in which this approach is based on simple approximations of the magnetization curve. After that, a torque ripple reduction is proposed, which are based on the adaptive identification of the equivalent torque coefficient. Finally, the performance of the proposed method is experimentally evaluated to show the feasibility of the proposed method.

To improve safety of wearable assist robots, the authors make a prototype wearable robot suitable for the contact force distribution feedback control system. Because wearable assist robots contact the user and apply force directly, ensuring safety of the user is most important. In order to improve safety from the aspect of control, a contact force distribution feedback control system has been researched. The system controls contact force between the user and a robot directly. Although the system validated in principal with a simple robot model on simulation, validation with an actual robot has not done yet. There are no wearable robots measurable the contact force distribution with user. In order to measure the contact force distribution for the control system, two points: measurement reproducibility and measurement as distribution, are necessary. In this paper, a prototype robot that satisfies the two points is made. The robot is confirmed to have the two points necessary for the system by measurement experiment.

This paper evaluates our measuring position determination method of Unmanned Aerial Vehicles (UAVs) for automatic visual inspection through an experimentation in real world. This method is under the assumption that the 3D model of a target facility, required accuracy of gathering data and specs of instruments are given. The triangular polygons for shape description of the 3D model are reconstructed for measurement and the measuring positions: the waypoints that an UAV travels to gather the accurate data of the target facility efficiently are calculated systematically. For basic evaluation of this method, we conducted an experimentation in real world. The measuring positions were calculated for an actual target and a camera. An experimenter brought a camera to the measuring positions and gathered image data. This experimentation result showed the effectivity of the method.

To realize automatic inspections for infrastructure facilities by an Unmanned Aerial Vehicle (UAV), improvement of localization performance is needed. Especially, automatic flight does not work in environments near facilities such as under the bridge because GNSS signals does not reach the UAV. We research a localization system using subsidiary UAVs that assist localization of an inspection UAV (main UAV). In this system, the positioning of the subsidiary UAVs is important to correctly complement the position information of the main UAV. In this paper, as the first stage of research, we consider a path of a single subsidiary UAV using preliminary information: the 3D geometrical data of a target facility, the inspection path of the main UAV, and the arrangement of GNSS satellites during the inspection. A simple simulation shows that the localization of the main UAV does not collapse through the whole path under the evaluation of the GNSS positioning precision of each UAV.

Sensorless control has become a popular research area for ac drives. However, most of the works are attributed to conventional three-phase drives. Given that applications of multiphase machines keep spreading, studies of position sensorless for them are necessary. For the object in this paper, some studies for sensorless control in middle/high speed region are reported. To accomplish a full-speed range sensorless control, a study about the standstill/low speed sensorless control of the target motor is proposed in this paper. Particularly, conventional signal injection methods utilized in three-phase ac drives meet a new issue caused by the inner magnetic couplings of the system when applying to the object in this paper. It may lead to failure of sensorless control and torque ripple enlargement. Therefore, analyses to solve the issue become the main interest of this paper, simulation results are demonstrated in the last section.

For the discussion about the neutral point potential (NPP) control of single phase three-level rectifiers, most of the methods solve the NPP problem by adjusting the zero-sequence voltage under symmetrical load conditions. However, when one phase voltage reaches to the maximum voltage, the other phase voltage goes to the minimal voltage. As a result, the NPP control ability is limited since there is no chosen zero-sequence voltage when the modulation ratio is close to 1. It can not be suitable for some strong asymmetrical load applications, in which two DC-link capacitors are used to drive different loads separately. In this paper, a dual-degrees of freedom based PWM is proposed, which can enhance the NPP control ability for single phase three level rectifiers. Introducing a new degree of freedom, the proposed method has more choices for the NPP control. It can decrease the NPP fluctuation under the strong asymmetrical load conditions, compared with the traditional PWM methods. The simulation results ratify the correctness and feasibility of the proposed PWM method finally.

Wearable assist robots deliver power in form of direct contacts with the user, therefore, safety control is considered to be the prior requirement for wearable assist robots. To improve safety, the robot needs to be controlled based on the current contact state feedback. As a state estimation concept approach, a feedback control system based on joint torque has been proposed. The principle of this approach is that the physical contact forces appear as a certain change in measured joint torque. However, as the joint torque is the resultant of contact forces, the case of user is in danger contact state may not be detected. In this paper, a new control system based on pressure distribution is proposed. Utilizing the pressure sensors directly mounted on the robot's link surface, the robot is controlled to suppress excessive contact forces. On computer simulations, the proposed system was confirmed to be effective to improve safety.

This paper presents a measuring position determination method of Unmanned Aerial Vehicles (UAVs) for automatic visual inspection. Distinguishing measurable geometry of a target facility by an UAV and ensuring reliability in data gathering to detect anomaly are both important functions for inspections. The proposed method is based on the premise that the 3D geometrical data of a target facility and feature of an UAV and a sensor are given. The measuring positions to gather the visual data ensuring reliability are calculated considering positioning error of an UAV systematically. By setting the measuring positions as waypoints, flight paths appropriating for visual inspection can be obtained by using conventional path-planning method. The proposed method is examined through computer simulations on the assumption that an UAV measure image data of several bridges in an actual environmental.

The authors have researched on a method of human action modeling to realize systems such as to support human human operations or watch persons to prevent various kinds of accidents. In order to recognize or support various kinds of human actions, a certain human action model is necessary in these systems. Therefore, we have proposed a modeling method of human actions, which is extracted statistically from enormous data acquired from various kinds of sensors by long-term observation of human actions and situations around persons. However, human action elements composing a model should have been extracted heuristically by a designer. In this paper, an extraction method of human action elements is proposed in order to extract frequent human action elements automatically from acquired data. In the proposed method, a series of acquired time series data is divided into short partial ones, which are modeled by Hidden Markov Models(HMM). Then, the transition network of generated HMMs is constructed based on the likelihood between the original data and each HMM. In the obtained network, a transition sequence with the only one edge is regarded as a frequent human action element. Extraction results with artificial and actual human action data are shown in this paper in order to verify the usefulness of the proposed method.

The traditional maximum torque per ampere (MTPA) control and the position-sensorless control based on the extended electromotive force (EEMF) with f-axis signal injection in zero/low-speed regions become difficult for low voltage high power permanent magnet synchronous motors (PMSMs), of which dq-axis inductances are susceptible to operating conditions. Taking full account of the motor particularity, it is pointed out in this paper that the dramatic changes in dq-axis inductances will lead to the incorrect orientation of maximum torque reference frame (f-t reference frame) and the failure of MTPA control. Thereby, the EEMF will be inadequate for the position-sensorless control. For the disadvantages of the prior methods, a novel position-sensorless control of low voltage high power PMSMs in zero/low-speed regions is proposed in this paper. By dealing with current control and position observation in different reference frames separately, the MTPA control and the position-sensorless control are well realized. The simulation based on JMAG model verifies the feasibility of the new proposal.

In this study, we aim to evaluate a driver skill based on the driver behavior model generated by the operation and environment data under driving vehicle. When skill of driver is reduced, the operator behavior is delayed, it takes along time of operation, and execution timing of operation is no reproducibility. These mean that reducing skill of driver affects time information of the driver operation. Therefore, we study on the model structure incorporated time information expresses the execution timing of driver's operation to evaluate driver's skill. In particular, we propose a new model structure incorporated timing probabilistic distribution expresses the time until the execution of next operation. Moreover, we propose a modeling method of operator behavior according to proposed model structure. It is considered that the reaction time from the decision to execution of operation and ambiguity of the decision can be read from the timing probabilistic distribution. In this paper, the above usefulness is examined through some experimental results. Moreover, the applicability of the driving skill evaluation based on the proposed operation model is examined.

This paper proposes a current control system using inverter overmodulation range by applying voltage phase reference filter for Interior Permanent Magnet Synchronous Motors (IPMSMs). Utilizing the overmodulation range can expand the operation range of IPMSMs with same battery unit. However, operation range cannot be expanded appropriately without countermeasure against harmonic components. The proposed control system can improve the current control performance by 1) applying the filter to the voltage phase reference and 2) varying the filter utilizing rate by the proposed filter switching algorithm without any motor parameters.

This paper describes current control system of PMSM for seamless drive between PWM mode, overmodulation and Square-wave mode of inverter based on model predictive control (MPC). Whereas typical current vector controller determines the ratio of voltage, our proposed current controller based on MPC selects directly the voltage vector sequence of voltage source inverter by prediction of the current behavior. It can be realized not only widening the drive range but also improvement of transient current response by utilization of aggressive square-wave mode in any operating point by selecting the sequence which has overmoduration and Square-wave mode. In this paper, we discuss implementation of proposed MPC and show the experimental results.

This paper concerns the sensor data fusion method in mobile robot localization. Sensor data fusion methods using particle filter have been a major technique to estimate robots's state from noisy sensor observation. Generally, the sensors assumed to be in the nominal state of work. In realistic contexts, however, sensor characteristics may change online depending on sensor functioning conditions, and deteriorate estimation accuracy. To generate robustness to disturbance, faulty sensors should be identified and the fusion rule should be updated accordingly. These processes should be done online without access to ground truth, known as Fault Detection and Isolation (FDI) problem. The major approach to FDI problem in mobile robot localization has been the model-based online monitoring of the sensor state utilizing sensor fault models. However, comprehensive theoretical models for complex systems are difficult to obtain, and in some situation impossible to derive. Therefore, authors proposed a new sensor data fusion method for mobile robot localization adopting the idea of majority voting FDI techniques, which has been a proven technology in the field of safety critical control, utilizing the increasing and diversifying onboard sensors thanks to the technological development in recent years. Under the strong assumption of majority voting that the majority of similar signals provide the truth and any dissimilar signal is the result of a sensor fault, faulty probability distributions computed from sensor observations by a particle filter are detected and isolated utilizing K-L Divergence, a measurement for distance between two distributions in information theory. The proposed method is examined on a computer simulation, showing the possibility of generating robustness to the sensor data fusion in the mobile robot localization.

In order to give a suitable support to a person timely, it is necessary for the system to estimate the next human action and its execution timing. Therefore, we have proposed a modeling method of human actions based on the causality between the situation around a person and human action change. Our previous modeling method makes it possible to estimate the next human action and its execution timing based on Hidden Markov Model which expresses the situation around a person. However, Hidden Markov Model was not enough to express the temporal information such as execution timing of next human action. According to this reason, we propose a new model structure of human action that a probabilistic distribution of timing information is incorporated into Hidden Markov Model. And we propose a new modeling method of human action based on the above new probabilistic model.

In this paper, a position sensorless control method for the wound-field synchronous motor with double three-phase wound stator in stop/low-speed driving is proposed. High-frequency signal injecton method which is commonly used for PMSMs is selected for the test motor's position sensorless control. However, due to the signal injecton, magnetic interference occurs between motor stator's windings. As a result, a large torque ripple generated on the output torque. Therefore, in this paper, using the feature of the test motor which has two sets of three-phase windings, the proposed method not only can realize the position sensorless control but can also reduce the torque ripple by injecting signals to the two sets of windings. Then, the effectiveness of the proposed method would be shown by simulation.

This paper proposes a direct rotor-position estimation method for IPMSMs (Interior Permanent Magnet Synchronous Motors) in low-speed regions including standstill. In general, the rotor position is estimated indirectly by converging the position estimation error into 0 by the phase-locked-loop (PLL). The indirect rotor position estimation methods require the convergence time of position estimation error. On the other hand, the proposed direct rotor estimation method does not require a phase-locked-loop (PLL), which degrades the response of position estimation. The proposed method has possibilities of improving the position estimation response. This paper demonstrates the performance of sensorless control based on proposed direct position estimation.

This paper discusses a voltage limiter based on a maximum torque control frame for interior permanent magnet synchronous motors. The maximum torque control frame can separate a current vector into two components: t-axis and f-axis. The t- and f-axis represents torque and flux components, respectively. We have already shown the effectiveness of the t-axis current change for the maximum torque response control under the non-voltage-saturation conditions. On the other hand, the voltage saturation condition is not discussed enough. When a output torque reference changes transiently, a voltage reference amplitude is easily over a maximum inverter output voltage. Therefore, it is necessary to limit the voltage reference amplitude. A lots of voltage limiters have been proposed, so far. In this paper, we show a voltage phase angle that maximizes the torque response at first, and then we show the relation between the voltage phase angle and the maximum torque control frame. Furthermore, the effectiveness of the proposed voltage limiter was verified through simulations.

This paper describes frequency characteristic design method of torque control system of permanent magnet synchronous motor (PMSM) based on model predictive control (MPC). MPC controller selects directly one of the voltage vector of three-phase inverter by estimating and evaluating future states (torque, current) of PMSM instead of generating voltage reference as conventional one. The sequence of selected voltage vectors makes inverter output voltage appropriate on the transient state in addition to on the steady state. As a result, this system obtains fast current response. However, torque control system controlled by MPC become nonlinear and frequency transfer function of the system can not be induced. Therefore it makes hard to design in cooperation with conventional linear systems. In this paper, we revealed frequency characteristic of MPC torque control system, and proposed designing method of frequency characteristic.

This paper describes position estimation method for interior permanent magnet synchronous motor (IPMSM) using idea of pattern matching technique. High frequency signal injection methods are conventional method of sensorless vector control for IPMSM in standstill and very low speed region. Those methods are used when the value of phase inductance can be assumed to be a sinusoidal function of rotor position. Generally, automobile motors are designed to have high torque density. In such motors, the value of phase inductance would be complicated because of the magnetic saturation, which negatively affects the accuracy of rotor position estimation. To solve those problems, we have proposed in this paper a new approach for sensorless vector control of the IPMSM with heavy magnetic saturation. This approach, utilizes pattern matching. Experimental results are included to demonstrate the applicability of proposed methods.

In this paper, a new impedance control method for wearable assist robots based 011 pressure distribution is proposed. Conventional power assist systems could not realize high power assist because they did not consider about the physical contacts between the user and robot. Also, these systems assisted only for simple joints such as knee, elbow, limb and so on. I11 order to achieve higher power assist lor multiple joints such as shoulder and waist, it is desirable to consider the contact condition directly instead of torque that was used in conventional methods. The authors propose a position based impedance control system using pressure distribution between the user and robot. By using a pseudo torque calculated by pressure distribution, the impedance controller can be expected to reduce the dangerous contact pressure efficiently. I11 this paper, power assist 011 the multiple joints are assumed: forward-backward bending movements of the upper half part of the body. A structure of the assist robot is determined based 011 analysis of human movements. By applying determined robot model 011 simulations, the proposed impedance control method is evaluated by comparing contact pressure to conventional methods.

In this paper, a method of Permanent Magnet Synchronous Motor(PMSM) position sensorless vector control system in the inverter overmodulation range has been proposed. Operating PMSM in the inverter overmodulation range achieves fast torque response and widens high-speed drive region of PMSM. In the range, however, because of the effect from harmonic currents, a closed loop current control system become unstable further more because of harmonic voltage output from the inverter makes position estimation error and it destabilize sensorless control system. First of all describes mechanism of destabilization of sensorless control system in the overmodulation range. Second of all propose a method which stabilizes control system. Last of all the experimental results show that the proposed method widen high-speed drive region and archives fast torque response.

In this paper, a modeling method of human actions is proposed. This is because systems to assist human operations have been desired, which must have a certain human action model to recognize or support various kinds of human actions. In the proposed method, a human action model is statistically generated by extracted from enormous data obtained by long-term monitoring with sensors. Therefore, human actions can be modeled without previous knowledge. In addition, a human action model is generated considering the differences on not only spatial patterns but also temporal ones of human actions. This is because it is very significant to support human operations timely. In order to generate a human action model with the previous two features, a human action is modeled as a human action pattern expressed by Hidden Semi Markov Model(HSMM) and a If-Then-Rule in the proposed method. HSMM can model a time series data explicitly. It can also model the timing information of a change of a human action. In addition, the relationship between the situation and a human action is expressed by If-Then-Rule explicitly. Therefore, a human action model has high readability.

In this paper, a novel correction method for current control of permanent magnet synchronous motors (PMSMs) has been proposed. In the high speed region such as the voltage is saturated, the conventional current control system of PMSM does not support correct operation, because inverters cannot be considered ideal power converter as regards this region. It has been pointed out that this problem of system failure is occurred by the incomplete decoupling control. In general, the operation of the integrator of the PI current control system is stopped when voltage saturation occurs. In this context, the integrator holds the value it had at the instant when the limit value was exceeded, but this value is not necessarily appropriate for the efficient decoupling control. For solving this problem, we propose a novel integral values correction method. The proposed correction method achieves stable operation also in the high speed region.

Several position sensorless control methods of PMSMs(Permanent Magnet Synchronous Motors) using the full-order observers have been proposed. However, the designs of the full-order observers are difficult, because the error equations of the observers become one of the fourth order. On the other hand, the the extended electromotive force(eemf) model which can be applied to all synchronous motors included Syn-RMs(Synchronous reluctance motors) had been proposed by authors, for the position sensorless control of IPMSMs(Interior Permanent Magnet Synchronous Motors). A full-order observer based on the eemf model(full-order eemf observer) which can be applied to the position sensorless control of all synchronous motors included SynRMs(Synchronous reluctance motors) have been proposed by authors. However, the design of a full-order eemf observer is also difficult. Therefore, this paper proposes a design method of the full-order eemf observer with the objective of the robust position estimation against the velocity estimation error using H∞control theory. Moreover, a new design method of the more robust full-order eemf observer is proposed by restudying the error system of the observer, and the simulation results and the experimental results show that a new design method of full-order eemf observer is very useful.

This paper describes position estimation method for interior permanent magnet synchronous motor (IPMSM) using idea of pattern matching technique. The IPMSM possess a unique relationship between a set of phase inductances and rotor position, that make it possible to estimate rotor position in standstill and very low speed region, where EMF become no enough for position estimation. Up to the present, many sensorless control methods for IPMSM based on magnetic saliency have been proposed. However, recently IPMSM, which possess higher torque density, the sensorless control performance to be degrade. To overcome this problem, we have proposed a new approach for position sensorless control, which utilize pattern matching. This paper reports of measurement of inductance spatial distributions in various load conditions and discuss feasibility of our proposed approach.

In order to realize a system to support human actions, the system should be able to estimate the next human action. Therefore, we have proposed a modeling method of human actions based on the causality between the situation around a person and the human action change. Our previous modeling method makes it possible to estimate the next human action. However, in order to give a suitable support to a person, it is necessary for the system to estimate not only the next human action, but also its execution timing. In this paper, we propose a modeling method of human actions for estimate the execution timing of the next human action by expanding our previous method. In addition, an early estimation method of the next human action and its execution timing is also proposed.

Model predictive instantaneous-current control (MPIC), which was proposed in our earlier works, enables us to achieve better instantaneous current control using mathematical models of an inverter and permanent magnet synchronous motors (PMSM). However, the dead-time to avoid the short breakdown in the inverter is the main reason in the modeling error. If the modeling error is not ignorable, it is not possible to predict the current evolution using the model. Thus, in this paper, we analyze the influence of the dead-time in MPIC and propose a compensation technique for the dead-time so that the error of the predicted current is reduced. The effectiveness of the proposed method is verified through simulation and experiments.

In this paper, a novel flux-weakening control method for Interior Permanent Magnet Synchronous Motors (IPMSMs) has been proposed. The proposed flux-weakening control method is discussed on the Maximum Torque Control (MTC) frame, which is possible to realize simple and robust sensorless control system. First, the proposed flux model for position sensorless control of IPMSMs has been reviewed, in which this model can approximately estimate the MTC frame, which stands for a new coordinate aligned with the current vector at the the Maximum Torque Per Ampere (MTPA) control. Next, a new flux-weakening method suitable for this flux model has been proposed. Finally, some experimental results demonstrate the effectiveness of the proposed method.

A new modeling method of human behaviors is proposed in this paper. In the proposed method, it is assumed that a person changes his behavior according to the change of the situation around him, and this concept is expressed by If-Then-Rules, which are called behavior rules. In behavior rules, a human behavior is described as a discrete event, and the change of the situation around a person is described by Hidden Markov Model (HMM) which models multi-dimensional time series sensing data. Moreover, and early estimation method of the next human behavior and the timing of its execution is proposed based on the proposed human behavior model. In this research, human operations of a radio controlled vehicle are modeled as an example of application of the proposed model. The usefulness of the proposed method is examined through experimental results of behavior estimation with the constructed behavior model.

We have proposed a modeling and recognition method of human behaviors in this research. In the proposed modeling method, we have assumed that a person changes his behavior according to the change of the situation around him, and this concept is expressed by If-Then-Rules, which are called behavior rules. In behavior rules, the change of the situation around a person is described by Hidden Markov Model(HMM) which models multi-dimensional time series sensing data. In this research, a support system for human driving behaviors has been assumed as an example of application of the proposed model. In order to realize the system which shows a driver his next behavior, we propose an estimation method of the next human behavior and the timing of its execution. The usefulness of the proposed method is examined through experimental results of behavior recognition with the constructed system.

We propose a modeling and recognition method of human behaviors in this paper in order to realize such intelligent systems that can adapt humans, i.e. the systems that support humans by considering human behaviors, In the proposed method, we assume that a human changes his behavior according to the change of the situation around him, and this concept is expressed by If-Then-Rules, which is called behavior rules. In behavior rules, the change of the situation around a person is described by multi-dimensional time series sensing data which is modeled with Hidden Markov Model(HMM). To recognize the change of human behaviors, the optimal If-Then-Rule is chosen based on the current human behavior and similarity to the time series data of the situation obtained by sensors. As an example of human behaviors, human driving behaviors are considered, and a recognition system of human driving behaviors is constructed. The usefulness of the proposed method is examined through some experimental results with the constructed system.

We suggest motion adaptation method against environmental changes for service robots. Several motions such as walking, dancing, demonstration and so on are described with time series patterns. These motions are optimized with the architecture of the robot and under certain surrounding environment. Under unknown environment, robots cannot accomplish their tasks. We propose motion generation system based on heuristic search with histories of internal sensor values. New motion patterns are explored under unknown environment based on self-evaluation function. Self-evaluation function is composed of difference of internal sensor values between standard environment and unknown environment. New motion patterns are generated to maximize self-evaluation function without external information, such as run length, position of robot, human observation and so on. Experimental results show that the possibility to adapt autonomously patterned motions to environmental changes.

Robot motions are designed according to the properties such as the number of joints, the range of each joint, the length of each frame and so on. Structure changes such as joint failures or frame distortion make the tasks of prepared motions unaccomplished. Identification of structure changes on multi degree-of-freedom (DOF) robots like humanoid is so difficult that the traditional approaches based on inverse kinematics would not be applied. We propose a motion adaptation method to generate new proper motions accomplishing the tasks on changed structure without identification of the structure changes. All the motions can be expressed with the time-series data of postures. Each proper posture as an element of motions for changed structure is estimated by Kriging with joint angle data of several original postures sorted by Vector Quantization (VQ) and the corresponding postures explored on changed structure by Simulated Annealing (SA). At the case the tasks couldn't be accomplished with the generated motions in designed DOF, new motions are generated in expanded DOF adding redundancy joints. Experimental results show that the generated motion by proposed method is accomplished 96.6% similarity with the original task referenced failure motion's task to 0% on changed structure by single joint locking.

The focus of this paper is mental tension detection in speech to assist control the tension in day-to-day business such as conferences and operations in a call center. It is difficult to use classical techniques for mental tension detection in dayto- day business because those techniques require invasion body by electrodes or squirts and tied up by cables. In order to achieve a non-invasive, non-contact and low-restricting method, this proposed technique uses acoustic features in the speech. The technique uses the vocal tract model which represents the shape and the tightness of throat muscle. The Gaussian Mixture Model (GMM) classifies two mental tension states: high-tension and non-tension. The experiment result shows high recognition rate of mental tension detection.

This paper proposes adaptive emotion recognition system in speech by feature selection based on KL-divergence. In order that the system can choose the most suitable feature set for emotion recognition, we propose an evaluation method for the set of prosodic features based on Kullback-Leibler divergence (KL-divergence). Additionally, we propose a method of feature selection system, using genetic algorithm (GA) making use of rapid evaluation based on KL-divergence. Experimental results show the proposed system can acquire efficient the prosodic feature set for emotion recognition in short order without constructing a recognition system. Furthermore, the accuracy of emotion recognition is significantly improved with the prosodic feature set selected by the proposed system.

Emotion expression through behavior is meaningful for communication processes between robots and humans. Automatic emotion expression generation is needed for those robots in order to stimulate communication with humans. Proposed technique generates emotional motions by modifying base motion patterns using the combination of the adjectival expressions. In order to generate emotional motions, emotional motion modifications should be based on rules which represent the relationship between emotion expressions and the combination of the adjectival expressions. Emotion modification rules are obtained by organoleptic tests. This paper, in particular, shows the relationship between the degree of adjectival modification and emotion expressions. Thus proposed technique generates the suitable emotional motion patterns from base motion pattern.

In this paper, we describe a motion pattern acquisition system from time-series posture data for a robot. In order to extract reusable motion patterns without the target pattern information a priori, proposed system extracts the patterns observed frequently. The extracted motion patterns often include transition motion at first and last. In order to reduce the transition part, we propose common block extraction method. The similar common block is classified to the same category. Then, the proposed system obtains the average of each cluster as representative motion patterns. Experimental results show remarkable motion patterns extracted by proposed system from time-series posture data.

Pulsed neural networks based on delta-sigma modulation (DSM-PNN) suitable for digital hardware implementation are introduced in this paper. Using delta-sigma modulation, which is a method to convert a signal into a 1-bit pulse-stream, it becomes possible to realize compact hardware implementation and accurate signal processing for pulsed neural networks. Since the proposed method has high versatility, various neural networks can be easily implemented in the hardware. In this paper, two pulsed neural networks, performing PCA or ICA, are introduced and their performance evaluated.

Neural network hardware is necessary for demanding real-time applications such as pattern recognition. In the paper, an improved calculation algorithm of the neuron model with sigmoid function is proposed, which is suitable for hardware implementation. The algorithm is based on the multi dimensional binary search. The neuron circuits implemented on FPGA by the proposed algorithm have shown excellence in size and circuit frequency compared with the conventional circuits with sum of product operation.