Published papers

© 1982-2012 IEEE. It is expected that bilateral control will find broad applications such as mobile robot exploration and operations in extreme environments. However, the transmission of vivid haptic sensation requires a rapid data-transfer rate of haptic data due to the short sampling period and high-frequency communication. The objective of this paper is aimed at the traffic reduction of haptic data by exploiting an impedance field equivalence of a standard bilateral control. The proposed approach theoretically clarifies the equivalence of the standard bilateral control and impedance control in terms of haptic transmission. Based on this equivalence, the same performance of a standard bilateral control can be achieved by only communicating the force information that determines the equilibrium position of the impedance control. The validity of the proposed approach is confirmed through both simulations and experiments. The obtained results show that the proposed approach successfully reduced the amount of communicated haptic data without any disadvantages compared with the standard acceleration-based bilateral system. This paper shows novel interpretation that haptic transmission is realized by controlling the equilibrium force, which contributes reducing data traffic in haptic transmission and understanding motion generation.

© 2018 IEEE. In recent years, the applications of robots have diversified, and interest in application to human living environment has increased. To co-exist with robots,it is necessary to ensure safety against collision. Conventional collision detection with a force sensor and collision mitigation by impedance control only have low adaptability. In thispaper, collision mitigation by combination of flexible structure and impedance control is proposed. Collision detection is performed by Load Disturbance OBserver (LDOB). In addition, switching control of the collision mitigation and damper control based on the estimated disturbance torque according to the environmental situation is proposed. Simulations are conducted forverifying the proposed approach.

© 2018 IEEE. In motor drive application, a high switching frequency is required in order to mitigate current harmonics which cause a copper loss. However, the high switching frequency makes the switching loss large. Hence, some switching loss reduction techniques have been studied before. However, they requires complicated circuits. This paper proposes a multilevel inverter topology for a switching loss reduction. By changing current paths, the proposed inverter realizes the zero current switching. It was experimentally verified that the proposed inverter realizes the zero current switching by a foundation circuit.

© 2019 The Institute of Electrical Engineers of Japan. In motor drive applications, a high switching frequency is required to mitigate current harmonics, which cause a copper loss. However, the high switching frequency makes the switching loss large. Hence, some switching loss reduction techniques have been studied in the past. These techniques have some problems. This paper proposes a multilevel inverter topology for switching loss reduction. Because of provision of multiple current paths, the proposed inverter realizes the zero current switching. Consequently, a loss of the whole system including an inverter and motor can be reduced with a higher switching frequency. The switching loss of the proposed inverter is smaller than that of the conventional one. It was verified that the proposed inverter has some advantages at high switching frequency, through calculations.

© 2018 IEEE. Motion transmission is a technology which is expected to be used for teleoperation systems. Sensorless control can improve fault tolerance of teleoperation robots exposed to harsh environments. However, sensorless control methods are generally sensitive to parameter errors of an actuator. A previously proposed sensorless motion transmission method especially requires the accurate value of resistance. Therefore, this work aims to identify resistance of two direct current motors used as a master slave system. Taking account of the master slave system of which common mode and differential mode correspond respectively to torque transmission and velocity synchronization, a least mean squares method jointly using a voltage injection is newly applied to the differential mode. Simulation results validated the proposed identification method and indicated that the sensorless motion transmission succeeded using the identified resistance value. The proposed novel method will improve the usefulness of sensorless motion transmission by adaptability to parameter errors.

© 2018 IEEE. The ambulatory gait monitoring systems (AGMS) using wearable devices have a great attention for health monitoring of individuals. Foot clearance is one of gait parameters and an indicator of gait quality and safety. Conventionally, foot clearance is calculated by post-processing using several methods such as extended Kalman filter (EKF), a weighted Fourier linear combiner (WFLC), a simple biomechanical foot model, zero velocity update (ZVU), and optimally filtered direct and reverse integration (OFDRI). However, real-time foot clearance (FC) estimation is required to apply to control of neural prostheses and assistive devices to prevent falling down. In addition, since humans adapt to several environments during gait motion, walking patterns are different among leveled walk, ramp walk, and stair walk. Therefore, environmental recognition also has an important role for the devices. Aim of this study is an implication of a system which performs real-time foot clearance and environment estimation in several situations. Different configurations of infrared (IR) distance sensors are utilized with a foot-mounted inertia measurement unit (IMU: an acceleration sensor, a gyro sensor, a magnetometer, and an air pressure sensor) sensor. Two IR sensors are attached to both sides of a shoe with different orientation. Developed system is tested in case of leveled walk, ramp walk, and stair walk. As the result, the developed system can estimate foot clearance and find characteristics of walking in different gait motion.

© 2018 IEEE. Technologies for machine automation are strongly demanded in some developed countries due to achieve various tasks which could not be treated by human. Autonomous Underwater Vehicles (AUVs) are currently being utilized for scientific, commercial and military applications. These vehicles require autonomous guidance and control systems in order to perform underwater tasks. Modeling, system identification and control strategy of these vehicles are still major areas of research and development. Various disturbance effects perturb a vehicle in underwater, but it is not really discussed by conventional control strategy. And thus this paper proposes a new control strategy which is still effective under the circumstances when the disturbance exists. Simulations were conducted for supporting the argument.

© 2018 IEEE. Machine automation is getting significant attention from the society due to the declining labor population in Japan. This paper considers Motion Copying System (MCS) that is known as the machine automation method. MCS is able to play back the saved motion if environmental conditions between the saving and loading phases are the same. The variation of environmental condition, however, degrades the motion reproduction performance because the tactile information needs the bilateral communication between the master and the slave, which can also be said of the sender and the receiver, unlike the visual or audio information. This paper proposes the scaling adjustment method for compensating the environmental impedance variation between the saving and loading phases. The effectiveness of the proposal was verified through simulation.

© 2020 IEEE. With the recent development of electric drive technology, systems with multiple electric motors are increasingly used in various applications. Among them, methods of driving multiple electric motors by using only one power source is studied. This multi-motor drive system contributes to the reduction of the number of power supplies and electric parts. Because of these characteristics, the system is effective in reducing the weight and the size of whole system. Therefore, this paper proposes a circuit that can control motors by frequency separation using the principle of amplitude modulation. As a result, it is possible to drive a plurality of motors independently and simultaneously with only one power supply. The simulations of multiple motor drives by the proposed circuit was conducted.

© 2020 The Institute of Electrical Engineers of Japan. Advanced rehabilitation requires a measurement system that quantitatively analyzes the actions of each individual. Joint moment is an indice used to quantitatively evaluate an individual's actions. Conventionally, force plates and highspeed cameras have been used to analyze joint moments. However, these devices need to be fixed in a laboratory, which limits the measurable motions and measurement range. Further, these sensors are expensive. Hence, it is difficult to use them in medical or nursing care facilities. Therefore, this study aimed to estimate joint moments using inexpensive and wearable sensors. This paper proposes a practical joint moments estimation system using inertial measurement units, which is a wearable sensor. The proposed method uses wearable sensors to measure continuous motion without limitation. It is also cheaper than conventional systems. The effectiveness of the proposed system was verified by comparison with the method using force sensors.

© 2019 IEEE. There are many patients with Parkinson's disease in Japan. The one of the initial symptom is tremor, which is involuntary oscillations of the body. Medications and deep brain stimulation are utilized as treatments for the tremor. However, side effects and the invasiveness to the brain are regarded as problems. Hence, the demand for the support equipment is increasing in order to suppress the tremor from outside the body. There are tremor suppression robots attached to hands and upper limbs. In other words, most researches focus on the tremor of arms. Therefore, this paper proposes the tremor suppression robot for a finger. Since the frequency of the tremor and the human intended movement are different from each other, the tremor suppression control is based on the frequency resolution. This proposed control achieves the following the human motions. The validity of the proposed method is confirmed by the simulation and experiment.

© 2019 IEEE. In this paper, Disturbance OBserver (DOB) and Reaction Torque OBserver (RTOB) considering reluctance torque is proposed. By changing the structure of DOB when performing Maximum Torque Per Amplitude Control (MTPAC), a high torque and a robust control system is designed. Specifically, a control system resistant to noise is designed by considering the reluctance torque in the torque constant part. The proposed method is effective in situations where noise is large, as in position sensorless control. Due to the noise resistance, the vibration of the current or torque is reduced. Also, by considering the reluctance torque, the estimated value of the reaction torque can be estimated accurately. Therefore the performance of the force control is improved. The effectiveness of the proposed method is confirmed by simulation.

© 2019 IEEE. This paper aims to realize the pitching vibration suppressing control for the Electrical Vehicle(EV). Over many years, the active suspension has been studied as an effective technology to suppress the vehicle vibration, however it requires additional parts and consumes the enormous electric power. This paper shows the vibration suppressing method utilizing the inertial force generated by the driving motor, and it achieves the low power consumption. Through realizing the control method considering the passengers' comfortability, Frequency-weighted Linear Quadratic Gaussian was expanded to Linear Quadratic Integrator system in this paper. To validate the effectiveness of the proposed method, the simulation has been carried out, and it was confirmed to suppress the vehicle body vibration.

© 2020 IEEE. Recently, attention has been paid to pushing task by robots. A flexible mobile robot with a spring was devised to achieve safe contact during the pushing task. In this paper, two types of pushing task were assumed, one is pushing a fixed object and second is pushing a mobile object. A control law for flexible mobile robots to perform these tasks was proposed. The first is resonance ratio control for force control on a fixed object. Second, deceleration time adjustment method was proposed to carry mobile object without overrunning the target position. Also an observer to estimate friction was proposed, which is an index for adapting deceleration time. The performance of the proposed methods ware verified by simulation.

© 2020, Fuji Technology Press. All rights reserved. With the remarkable development of related technologies, the number of robots has been gradually increasing and their presence is becoming much more familiar in our daily lives. The motion copying system (MCS) is utilized as the method for conducting some tasks by robots. This system enables tasks to be reproduced when the environmental conditions are not changed. The task reproduction performance is degraded when environmental variations occur, and human-like adaptable motion is expected to be developed in the MCS. This study reveals the dominant element of motion, and the control strategy is varied at each time in each axis by considering the task realization. The flexibility of motion is learned from both the operator and the task implementation. The task reproduction experiments by MCS are conducted to verify the effectiveness of the proposal.

© 2020 IEEE. This paper proposes a novel online motion modification method in motion reproduction system. Motion reproduction is used to reproduce human motion saved through bilateral control. Bilateral control is a teleoperation which can transmit both position and force. Human judgement is needed for compensating the saved motion when the environment is different. However, conventional motion reproduction system only relies on saved motion or environment in slave side. Therefore operators in master side cannot modify the motion after motion has been saved. Instead of just reproducing the motion data, the proposed method adds master system at motion reproduction phase so that an operator can modify the motion. If the master system is simply added to motion reproduction system, the operator's motion and saved motion will interfere each other. Therefore, we propose to use weightings to suppress the interference. Simulated results validates the effectiveness of the proposal.

This paper proposes the robust system to control the position, force, posture, and moment of the multi-link manipulators. Disturbance Observer (DOB) has been researched as an effective method for the robust motion control, however the rotational components such as posture and moment had not been considered in conventional studies. This paper expands the existing DOB-based translational position control to the position-posture control in 6 Degrees-of-Freedom workspace using Euler angle/Quaternion which are the representative methods to describe the posture. Furthermore, the position-force-posture-moment hybrid control is proposed based on the position-posture control. Through the numerical simulation, the effectiveness of the proposed method was confirmed.

The world population is ageing. In the aged society, grasping the body condition of the elderly is helpful to choose appropriate rehabilitation. This research focuses on the outdoor walking analysis by use of the simple foot mounted IMU system. The walking feature values are estimated by the algorithm, and they are combined with the GPS data obtained by smartphone to compare the feature values of frail elderly and healthy subject. The differences of walking feature values that are calculated by this system show the usefulness of developed system. The developed system is simple, but useful to grasp the gait features changing of the user.

© 2018, © 2018 Informa UK Limited, trading as Taylor & Francis Group and The Robotics Society of Japan. This paper aims at developing an estimation method of end-point impedance. Human operators are constantly changing their end-point impedance for adapting to the surrounded environment and executing some complicated tasks, and it is highly meaningful to investigate these variations for the further understanding of human motion. Most of the conventional researches, however, have considered non-contact-point impedance or tasks that is only holding the vibrated sticks due to the experimental constraints. This paper proposes the estimation method of end-point impedance by using bilateral control system. The extra signal is added to the force controller for the impedance estimation. In addition, the effect of the bilateral controller is estimated and removed from the impedance estimation process for securing the applicability of moving tasks. The proposed method was validated through simulations and an experiment. The experimental result showed that the end-point stiffness can be estimated properly even if the operator robot was moving and changed its end-point impedance.

© 2018 The Institute of Electrical Engineers of Japan. Road traffic accidents are a leading cause of avoidable deaths, especially among young persons. Because driver behavior is a major cause for these the incidents, any device able to help the driver should lower casualties. As an intermediary step between autonomous vehicles and human driven ones, active driving assistance technology can help the driver maintain a safe trajectory. Applying remote control to vehicles could also decrease the probabilities of accidents, by facilitating the reduction of the duration of the driving sessions, especially in fret transport applications, decreasing drowsiness of the driver, a factor that increases the occurrence of accidents. However, the addition of remote control conducts to a communication time delay that can be dangerous in cruising situations. This paper presents the application of an active driving assistant to compensate for the effects of the communication time delay in collision avoidance scenarios using a small-scale vehicle.