Published papers

This paper proposes an unscented Kalman filter (UKF) with disturbance estimation for an underwater quadrotor control system by utilizing backstepping control. Autonomous underwater vehicles (AUVs) have been attracted attention to scientific and commercial applications. The tasks in those applications are such as surveys and inspections for various objects underwater in narrow space. In this study, a quadrotor type robot, in which high-performance results are obtained for aerial application, is regarded as AUVs. The quadrotor robot is a smaller system than general AUVs and has suitable merits for work in a narrow place. On the other hand, since the quadrotor system is small and light in weight, it is more susceptible to underwater waves than conventional AUVs. Therefore, consideration of a technique to suppress or to reject the influences of the waves in the control systems is a must. This paper proposes a UKF system including a second-order model of input disturbance for estimating the influence of waves and an accurate quadrotor state simultaneously. Additionally, the disturbance estimation performance assists in make robustness as its estimated disturbance is rejected to actual disturbance. Finally, the usefulness of the proposed system is shown via simulations of position control for the underwater quadrotor affected by a wave effect model.

Human support robots are in high demand and the performance enhancement through force control has been extensively studied. However, the design of the force controller and selection of appropriate gains are sometimes difficult because they are affected by the conditions of motion such as environmental impedance or model uncertainties. This study discusses force control with a disturbance observer and applies a neural network (NN) into its controller; the NN works as both the feedback and feedforward components. The contribution of this study is to show the development method of force control using disturbance observer and a NN, which enhances the performance of force control from the perspective of both feedback and feedforward components. The structure of the controller and composition of the NN were selected through simulation results; moreover, the compensator based on NN was designed in a frequency range higher than the cutoff frequency of the observer with a small number of hidden layers. Moreover, this study discusses a training method of weights in real time. Simulations and experiments were performed for showing the effectiveness of the proposal.

This paper describes the implementation of a disturbance observer-based controller after Savitzky-Golay filters. The cutoff frequency is a bottleneck for the conventional disturbance observers in DC motors with position measurement. The measurement noise limits the cutoff frequency of the pseudo-derivative in the design of the conventional disturbance observers. This study proposes an implementation of a disturbance observer based on Savitzky-Golay filters to improve the performance in noisy conditions with respect to the cutoff frequency of the time derivative. The proposed method was verified through various numerical simulations and experiments, thereby demonstrating its applications in bilateral control.

© 2019 IEEE. Technologies for machine automation are strongly demanded in some developed countries for achieving various tasks which should be treated by the robot. Autonomous Underwater Vehicles (AUVs) are currently being utilized for scientific and commercial applications. To achieving these tasks, it is required autonomous the guidance and the control systems. Therefore, modeling, system identification and control strategy of these vehicles are still major areas of research and development. Normally, the full actuated AUV is considered for facilitating its control and the disturbance compensation. However, it increases the cost to create and maintain. And thus in author's previous research, an under actuated AUV which still can be controlled when the disturbance exists, was considered to decrease the cost with the same performance as the full actuated one. However, some disturbance effects could not be compensated because of severe nonlinearity of it and the non-holonomic constraint. Therefore, the new control strategy which is considered its nonlinearity, is proposed in this paper and the simulation supports this argument.

© 2019 IEEE. This paper deals with the decentralized force control problem of compliant robots driven by Series Elastic Actuators (SEAs). The decentralized force control problem of a compliant robot manipulator is transformed into the robust position control problem of a servo system by using Hooke's law and lumping the nonlinear dynamics and unknown disturbances of the robot manipulator into a fictitious disturbance variable. In order to precisely follow the reference force trajectories at joint space, i.e., the position trajectories of the servo systems, the robust controller is synthesized by using Disturbance Observer (DOb) and Sliding Mode Control (SMC). Although the robust force controller can be synthesized by using only SMC, the force control signal may suffer from high control signal chattering as the force reference input is increased. In addition to improving the robustness of force control, DOb significantly suppresses the discontinuous control signal chattering by allowing to decrease the SMC gain. When the proposed decentralized robust force controller is implemented, the compliant robot can precisely track desired force trajectories and safely contact to different environments. The validity of the proposed force controller is verified by giving the simulation results of a redundant robot manipulator.

© 2019 IEEE. Accidents caused by a collision between an environment and a robot become a problem, so safety strategies of robots are indispensable. Since the impact force sharply increases at the time of a collision, position-based impedance control which has low responsiveness to the collision can not sufficiently reduce the damage. This paper proposes the manipulator with passive and flexible base for collision force suppression. The proposed flexible base has a rotating structure and it is possible to suppress the collision from any direction when an environment contacts the working manipulator. When a large external force is applied, the variable impedance control according to the strain amount of the base spring in order to avoid the strain limit of flexible base is proposed. Also, when the external force is estimated with force sensorless, the estimation error increases due to the friction model error of the motor, and the position followability deteriorates. Therefore, the contact state is judged by the strain amount of the flexible base, and correction is added to the estimated reaction force, thereby improving the position followability of the impedance control in non-contact.

© 2019 The Institute of Electrical Engineers of Japan. This study aims to realize an automated deceleration system to avoid the collision of a vehicle with a pedestrian. The proposed system can predict a pedestrian’s future position from the current position, and detect the collision probability. Furthermore, the controller employing model predictive control, which can compensate the driving resistance and the modeling error is also proposed. The effectiveness of the proposed method is verified through a simulation and experiment.

© 2019 IEEE. Nowadays, a low birthrate and longevity are proceeding in developed countries. Due to the low birthrate and longevity, there is a shortage of caregivers. Therefore, an assistance of nursing care is necessary. Also, according to a multidisciplinary advance of robots, robots increase opportunities for operating in environment in which human intervene. Operating a robot in human presence is a dangerous for human. Therefore, a signal transmission between human and robots is necessary. This paper uses a cuff of sphygmomanometer as an interface of signal transmission. This interface is attached to an arm of user and an air pressure of interface is controlled to constant. An error between pressure command and response caused by a muscle bulge of user is regarded as an intention of user. An integration of error is given to a manipulator as a position command. According to this system, the interface user can operate a position of manipulator based on the user's intention.

© 2018 IEEE. Along with a development of the industry, wearable devices are spreading. These devices often have a problem of a battery exhaustion. Biomechanical energy harvesting is one of the methods to solve this problem. As methods of biomechanical energy harvesting, the shoes which attached piezoelectric elements and the system which attached a generator to the knee were researched.This paper focuses on the surplus energy when walking down a slope. In order to determine the amout of the surplus energy, a passive walking model is utilized. Additionally, a method to recover surplus energy while maintaining walking stability is introduced. Finally, PI control of the boost chopper is utilized to adjust a braking torque and the amount of the generation. The validity of the proposed method is confirmed by simulation.

©2020 The Institute of Electrical Engineers of Japan. A bilateral control system aims at realizing a fine haptic transmission between a human operator and a remote environment object. Such a system is expected to realize multiple requirements in practical applications, and the architecture design based on a layer structure has to be considered. This paper treats a surgery robot system as an example, and its system has two requirements: "large workspace"and "fine haptic sensation". Using the conventional bilateral control system to realize these requirements simultaneously is difficult because mechanical and control designs are separately discussed. At the same time, a system that has multiple requirements such as a fine/coarse system is often designed only using position information. This study proposed to consider not only mechanical/control design but also the position/force information simultaneously from the perspective of an architecture design. The proposed architecture decomposes the functions into each motor/robot by defining roles and combining each role in the layer structure. The advantage of the proposal is that a complex system can be designed as a entire by combining multiple sub-functions that can be simply designed. As a result, the surgery robot system is designed by considering its requirements, and the experiments are carried out to verify if the requirements are satisfied. Â

© 2020 Japan Society for Precision Engineering. All rights reserved. Teleoperation is one of the technologies for supporting operators, and its development is strongly demanded. Acceleration-based four-channel Bilateral Control (ABC) is known as a method of teleoperation and achieves highly precise impedance transmission. The gain scheduling for bilateral contrai has basically been focused on the stability by the model-based approach. However, describing the contact motion and precise identification of some nonlinear terms are sometimes highly difficult. Moreover, the gain parameters between the master and slave tend to be set the same although the transfer function of them are different. Therefore, this paper considers the gain setting from the perspective of precise impedance transmission and proposes the evaluation index-based gain scheduling for A B C by using the experimental data. This paper also discusses the designing of gradient-descent algorithm for an actual gain setting by taking the obtained knowledge from evaluation index-based gain scheduling into consideration. The effectiveness of proposed gain scheduling and gain tuning methods were verified through experiments.

© 2020 IEEE. The aim of this study was to stabilize the posture of a two-wheeled walker in a torque saturated situation during turning. A two-wheeled walker increases the torque of one of the wheels during a turn, which may result in the torque saturation. There was a problem that the posture stability could not be maintained due to this torque saturation. Therefore, this paper proposes to generate a torque reference value in a torque saturation situation when a two-wheeled walker is turning. The simulation in the proposed method was compared with the case without considering the torque saturation. The simulation results showed that both the posture stability and the turning performance were improved. This indicates that the generation of the torque reference value considering the torque saturation is effective. And, we also propose a method to make the two-wheel walker stand still.

© 2020 IEEE. With the rapid aging population, the need for standing-up assist device has grown because standing-up movement is essential for independent living. The standing-up assist controller needs to accomplish high trajectory-following without high rigidity. Therefore, compliant force coordinate transformation control (CFCT) is applied to the controller. This control method induces user to model trajectory effortlessly. Moreover, this paper proposes the algorithm which considers the differences of standing-up movement speed. The proposed algorithm enables the assist device to support the user regardless of the difference of movement speed. The controller is evaluated through experiments.

This paper proposes a robust hybrid position/force control (HPFC) system for multi-degree of freedom manipulators (MDoFMs) with torque constraints on each joint. General HPFC systems can control the interactive contact forces and positions of manipulators in various environments. In HPFC systems, to improve the control performance for MDoFMs, motor torque saturation should be considered. Thus, a robust HPFC system that considers torque constraints by predictive functional control (PFC) is proposed in this paper. The proposed method simultaneously handles the response characteristics and torque constraints of actuators by using PFC as joint space position controllers. Additionally, the robustness of actuators against external forces is enhanced, and the model parameter errors are compensated by the disturbance observer technique. Moreover, implicit force control law and inverse kinematics are introduced for the workspace position/force control to implement the joint space position controllers. Consequently, the stability and robustness of the actuators are considered, and suitable position/force control can be performed even if the torque is saturated. The validity of the proposed method is tested with three planar manipulator joints with a uniaxial force sensor.