論文

The quiet standing test is used to detect diseases of the postural control system. The descriptive statistics of the center of pressure (COP) of older people during the test tend to be larger than those of healthy young people, but they cannot indicate structural problems in postural control. COP trajectories can be mathematically modeled with structural parameters such as viscosity, stiffness, and stochastic terms; however, the classification accuracy of older and fall-experienced people using such parameters has not been sufficiently verified. In this study, six structural parameters of a mass-spring-damper (MSD) model were estimated using two datasets, in which a total of 212 subjects performed quiet standing tests under four conditions. The estimated parameters were used for classification with a random forest algorithm to examine the differences in classification accuracy compared to seven conventional descriptive statistics methods. For the classification of older subjects, the classification accuracy of the MSD parameter method was the highest in foam condition, with positive likelihood ratios approximately 8.0. For the classification of fall-experienced subjects, the positive likelihood ratio of the MSD parameter method was 5.0, which is better than conventional descriptive statistics. Various MSD parameters revealed that aging and changing the floor surface and visual conditions cause oscillations in the COP behavior. While the MSD parameters were confirmed to help classify older subjects more accurately than the conventional descriptive statistics, there was room for further improvement in the classification accuracy of fall-experienced subjects.

This paper presents a control method with an estimation method of cross-coupling factors of Interior Permanent Magnet Synchronous Motors (IPMSM) inductance for position-sensorless servo drive systems. The cross-coupling factors are from dq-axis mutual inductance. Conventionally, the cross-coupling factors have been ignored in the IPMSM model and deteriorate control performance. That has caused decreasing robustness and velocity vibration in the low-speed range. Since the cross-coupling factors vary with the rotor position and the current, it is hard to measure them. Therefore, the proposed method utilizes a high-frequency voltage injection method to estimate the cross-coupling factors online. In addition, the proposed current control method utilizes a sliding mode control and voltage disturbance observer based on the voltage equation, including the dq-axis mutual inductance model. The estimated cross-coupling factors are reflected in the proposed current controller instantaneously. Experimental results show that the proposed method considering cross-coupling factors increases the robustness to the disturbance force and reduces the velocity vibration.

Complex slippage and nonholonomic constraints can disturb the precise movement of tracked vehicles. In this study, a position control system using driving force control and virtual-turning velocity control (VTVC) was established for a tracked vehicle. Slippage in the translational direction can be suppressed by controlling the driving force of the crawlers, estimated using the vehicle velocity. However, the driving forces interfere with each other in a turning motion. Therefore, a driving force distribution method was developed using the instantaneous turning center of a vehicle to decouple the forces. This distribution induced slippage in the direction of turning. In addition, a virtual reference was derived from the vehicle velocity and lateral disturbance, which described the effects of the nonholonomic constraints and skidding of the tracked vehicle. Next, a VTVC method was developed to suppress the lateral disturbance by controlling the turning velocity to follow the virtual reference. The experimental results confirm that the proposed approach ensures high-performance position tracking and adequate slippage of the tracked vehicle.

Elderly and disabled people frequently experience falls that may require early assessment and training. The quiet standing test provides descriptive statistics such as the center of pressure (COP) sway data, which can be used to analyze the decline in balance ability. Generally, higher descriptive statistics indicate lower balance ability. Stochastic models can model COP trajectories, and such equation parameters were previously used to assess the balance of patients. However, the model equation is a hypothesis and should be verified. In this study, we evaluated whether stochastic models can predict the descriptive statistics of observed COP trajectories. We estimated the model parameters by fitting postural sway data from 49 individuals in four linear stochastic models, and the prediction accuracy was verified by comparing the observed descriptive statistics with the predicted COP trajectories. We observed that the prediction accuracy of the models with stiffness was much higher than those with viscosity only, and the models with the center of mass (COM) had higher prediction accuracy than COP-only models. Therefore, we identified that coefficients from Langevin-based models contained descriptive statistics about a subject's COP trajectory, which suggested that these coefficients can be used to effectively assess balance issues.

Quantifying walking stability is needed in order to evaluate the various walking assist methods. It can also allow the planning of individualized interventions for elderly people or patients with disorders. The Gait Sensitivity Norm (GSN) is a new measure of the walking stability developed for passive dynamic walkers and has been reported to show a strong correlation with the fall risk. The GSN is a method based on the idea of the H2 norm, which allows us to know the dynamic response of the system of walkers by measuring the impulse response. The reliability of the GSN quantification has been confirmed in passive dynamic walkers and neuromusculoskeletal walking models. However, it has not yet been applied to humans. In this paper, we proposed a method to apply the GSN to the analysis of human walking. To apply the GSN to a human gait, wearable sensors were used to analyze human walking. The validity of quantifying gait stability with the GSN is assessed in the experiments comparing abnormal walker (with some weight or blindfold) and normal walkers. The result is consistent with the intuitive idea that blind walking is more unstable than normal walking.

Sensing is essential in human cooperative robots to realize safe and powerful operations against contact with the surrounding environment. Contact sensing has been studied for manipulators to avoid accidents caused by collisions between humans and robots. Conventional research has demonstrated that contact sensing requires many sensors or has a limited measurable range. In the conventional research without sensors, modeling systems, such as frictional force and plant parameters, have been used to estimate the contact force. However, a high reduction ratio mechanism for achieving powerful operations hinders accurate force estimation. This study proposes a method to estimate the contact force and contact position on a manipulator with a high reduction ratio mechanism by installing only one force/torque sensor on the base frame of the manipulator. By using the proposed method, the contact force and contact position can be detected regardless of the location where the contact force is applied. Furthermore, when the contact force is applied in a location where the contact position cannot be accurately estimated, an estimated disturbance or a link motion is used to determine the contact position. The effectiveness of the proposed method is validated through experiments.

This paper proposes a bilateral control system implementing an admittance control system. Bilateral control system realizes teleoperation of a remote robot with a sense of touch. We focus on an admittance-based bilateral control system, which needs only two communication channels and has almost the same control performance as an acceleration-based bilateral control system, which is one of the highest transparent teleoperation systems. However, the conventional admittance-based system assumes an ideal communication situation where there is no communication time delay and no packet loss and has possible to increase position error as time goes on in practical use. The proposed admittance-based system achieves the same control performance as the conventional structure without increasing position error even if there is a communication delay. The validity of the proposed system is confirmed through simulations and experiments.

This paper addresses the influence reduction of quantization and observation noises in a disturbance observer (DOB) technique. DOB is a disturbance estimation method that makes control systems robust. However, in implementing low-resolution sensors, disturbance estimates from DOB are considerably influenced by observation and quantization noises. In this paper, a novel DOB design method for simultaneous estimation of state and unknown disturbances, including the reduction of noise influences, is proposed. The proposed method is divided into two components. The first component is a Kalman filter (KF)-based DOB for simultaneous estimation of state and unknown disturbances. To improve the estimation performance through the KF-based DOB, a forgetting factor-based adaptive KF (FAKF) was employed. The second component is an adaptive law for the forgetting factor in the FAKF. The adaptive law is used for balancing the estimation accuracy and observation noise reduction. Simulation results involving various types of noise environments demonstrate the effectiveness of the proposed method.

© 2018 IEEE. Recently, Wireless Power Transfer (WPT) via Magnetic Resonant Coupling is especially focused around the world. Since a space will have a lot of devices in the future, WPT considering multiple devices will be required. It is important to consider not only power transfer efficiency but also power distribution when power is wirelessly transferred to multiple devices. The amount of received power in a receiver depends on the impedance of the receiver and the ratio of power distribution has the relation to the impedance. Conventional WPT system requires a few controllable repeaters and the large size of the receiver. In this paper, a variable impedance circuit for power distribution is proposed. Using the proposed circuit controllable by DC voltage, the problem for the conventional system is solved. In the future when WPT is popular, the circuit realizes electrically controllable power distribution.

© 2019 IEEE. Sensorless bilateral control has been studied to improve fault-tolerance of teleoperation without using torque sensors, position sensors, and other mechanical sensors. However, sensorless control often has less control accuracy due to model errors, parameter variations, and other unexpected factors. Feedback control with some position sensors is practically required for teleoperation. Therefore, this study aims at sensorless bilateral control using Hall ICs. A proposed bilateral control does not necessarily require disturbance observers unlike a conventional method originally using high-resolution sensors. Owing to this advantage, low-resolution sensors are applicable to the proposed bilateral control. Experiments showed that the proposed method succeeded bilateral control using Hall ICs without mechanical sensors, although the conventional control did not work using the low-resolution sensors. Therefore, the proposed method will be able to enhance fault-tolerance of teleoperation systems.

© 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group and The Robotics Society of Japan. This paper proposes a novel, simultaneous bipedal locomotion method using haptics for remote operation of biped robots. In general, traditional biped walking methods require very high computational power and advanced controllers to perform the required task. However, in this proposed method, a master exoskeleton attached to the human’s lower body is used to obtain the trajectory and haptic information to generate the trajectory of the slave biped robot in real time. Lateral motion of the center of mass of the biped is constrained in this experiment. Also, it is considered that no communication delay is presented in between the two systems in this experiment, and they are not discussed in this paper. Since a direct motion transmission is used in the proposed method, this method is quite straight forward and a simultaneous walking can be realized at the same time with high performance. Also, it does not require an exact dynamic model of the biped or specific method to plan the trajectory. The gait pattern of the biped is directly determined by that of the human. Also, the operator can feel the remote environment through the exoskeleton robot. Results obtained from the experiments validate the proposed method.

© 2018 The Institute of Electrical Engineers of Japan. An aging society is a serious problem in many developed countries. Due to this issue, there has been a significant increase in the demand for assist devices that can support human motion. This paper proposes a novel assist device, Intelligent Tension Pole (ITP), to assist in indoor activities of daily living (ADL), and it is expected to provide support for achieving self-reliance, space saving, and stability. The structure and control system of this assist device are presented. This novel assist device is based on a movable handrail, and therefore wheels are added to the lower and upper ends of the tension pole. In addition, a ball screw, which enables the ITP to move while stretching to the floor and ceiling at any time using force control, is added. In order to improve user comfort and reduce the risk of falling down of the ITP, translation and inclination controls of the ITP are applied to the wheel control. These controls are controlled independently by the mode decoupling method. Experiments are conducted to verify the effectiveness of the proposed device and its control system.

© 2019 IEEE. Teleoperation gets a significant attention from the society, and 4ch bilateral control is known as a method to realize the haptics transmission. Its controller is able to achieve the rich haptics transmission, but the performance is degraded under the presence of time-delay due to the interference between the position and force controllers. In particular, its interference becomes larger when position and force scaling are set more than 1 and the mass difference between the master and slave is large. 3ch architecture has proposed to tackle with the performance deterioration when the position and force scaling are set more than 1, but no researches have conducted for considering both the scaling and mass difference. This paper considers the situation that the scaling are set and the mass difference is existed in addition to the time-delay. Oblique coordinate control is utilized, and the superiority of 4ch and 3ch architectures are discussed. The proposal takes the superiority of these two architectures, and its effectiveness was verified through simulations and experiments.

© 2018 IEEE. Gait motion is the essential function for human being. Some diseases make it difficult for people to walk. Walking assist robot has been developed to assist gait motion for disabled people. Knee Exoskeleton is one of the walking assist robots. Knee Exoskeleton generates the assist torque for knee flexion or knee extension. The direction of the assist torque is controlled by the gait phase detection method. However, the conventional robot interferes with human intention because it generates flexion torque consistently in stance phase. In early stance phase, the user wants assist torque to support knee extension, but he wants knee flexion assist torque in late stance phase. The conventional controller does not detect late stance phase. In the previous study of the author, the novel gait phase detection algorithm is proposed to distinguish late stance phase. In this research, the novel algorithm is introduced for the proposed controller. A subject walks on the treadmill while wearing the knee exoskeleton. In the experiment, the exoskeleton is controlled by the conventional controller and the proposed controller. EMG of the popliteal muscle during gait motion is analyzed to evaluate whether the proposed controller succeeds to prevent interference with user's intention. Through the experiment, it is verified that the proposed controller succeeds to decrease the EMG of knee muscle by distinguishing early stance phase and late stance phase.

© 2020 IEEE. Recently, two-wheel mobile robot such as Segway is expected to be applied to a personal mobility. And it has the advantage of easy turning and stepping over and the disadvantage that many sensors are required for stabilization. In conventional researches, the stability and operability of the two-wheeled mobile robot were improved. These are established by assuming that 'the wheels do not slip'. However, a slip of the wheels cannot be ignored in actual driving, so that it is difficult to stably travel on a slippery road. In this paper, we consider the body-velocity by assuming 'the wheels slip in the modeling'. Sliding mode control system by using the body-velocity was proposed. Then, the acceleration observer is introduced to compensate the acceleration on body velocity. We consider both the body-velocity tracking and the pitch angle stabilization and achieve stable traveling control on slippery road. The validity of the proposed method is confirmed by simulation results and experimental results.

A stability index has been generally used in a gait system of formulaic structure. In this standardized structure, a human model has a center of gravity (COG), while a robot device has a zero moment point (ZMP) with COG as the stability standard in the walking motion. In the gait system of wearable type, the stability of human model is closely related with the robot device, in other words, the COG and ZMP are located in the limited domain of human motion and robot task. In this paper, the human and device are designed individually and the knee joint of human is unconstrained from the device work to allow the human reliance. The stability index that includes the COG and ZMP is defined. The validation of this index is guaranteed through a few tests to describe a correlation between the stability index and the human model condition. Based on the validity of the stability index, the experiment for the walking stabilization is implemented with the ZMP estimation by reaction torque observer.

© 2018 The Institute of Electrical Engineers of Japan. This paper aims to realize position and torque sensorless motion transmission between a generator and a motor. The proposed voltage compensation method requires only one current sensor on the generator side. Since the motor side is completely sensorless, this system is suitable for operation under harsh environments. Comparative analyses indicate that the proposed method satisfies the conditions for the transmission of an actual tactile sensation. Simulations and experiments confirmed that motion transmission is realized using the proposed method.

© 2019 IEEE. Most countries are running shortage of working force due to the aging population and reduction in the birthrate. Robot manipulators are expected to replace human work. However, it it still difficult for manipulators to do simple tasks such as fruit harvesting, foods cooking or toy assembling. A problem for robotic automation arise in the difficulty in teaching how much force manipulators should use for a task execution. Motion reproduction system, which uses bilateral control to store motion data, is one of a method to teach manipulators motion including position and force. The problem concerning motion reproduction system is that the motion reproducing fails if environment is changed between motion saving phase and motion reproducing phase. Motion reproduction system which can understand and adapt to environment is required. Vision sensor can sense environment. Computer vision is mainly focus on how to classify objects. Vision information is seldom combined with motion control especially force motion. Therefore, I propose a motion reproduction system in which reproduced motion is decided based on several motions and collected depth data. Convolutional Neural Network(CNN) was used to estimate a motion command from a depth image. Saved force data was used to generate labels for training. The label decision is different from conventional Machine learning alzorithm.