Musculoskeletal humanoids possess flexible and redundant bodies that closely resemble the human body. However, their application in real-world tasks has been impeded by the challenges involved in achieving bipedal walking with flexible bodies. To address this issue, we developed Musashi-W, a musculoskeletal wheeled robot. The Musashi-W comprises an upper body musculoskeletal system, lower body wheels, a linear motion mechanism that corresponds to the back, and a rotatable head equipped with a camera. Nevertheless, managing the combination of different circuit systems and drive systems has been a challenge. Hence, we developed an integrated infrastructure capable of managing these systems seamlessly. Moreover, body schema learning, reflex control, and the utilization of variable stiffness are crucial for handling musculoskeletal humanoids. We have combined these elements to construct a task realization system using musculoskeletal humanoids. We demonstrated the effectiveness of our system by realizing a table-setting task through dynamic cloth manipulation using variable stiffness.

This work investigates the feasibility of a novel decontamination technology that employs ultrasonic acoustic streaming to improve aerosol distribution inside isolator chambers. For this purpose, a numerical model consisting of three simulations (i) acoustic pressure, (ii) fluid flow, and (iii) particle tracing, is developed. Acoustic air streaming is examined for ultrasound source pressure levels between 131 and 136 dB. Maximum streaming velocity reached values of the order 10−3 and 10−2m/s which agrees with previous experimental and numerical investigations. The effect of air velocity field on the deposition of the decontamination agent on the isolator floor was assessed for aerosol with droplet diameter between 10 and 20μm. The results show that a more homogenous distribution of deposited droplets is obtained with configurations that have sound sources placed at a distance from each other. Designs with a homogenous aerosol deposition are expected to have a better decontamination outcome.

Humanoid robots are expected to expand their activity locations to narrow or uneven environments. When robots operate in a narrow space or on uneven terrain, they receive unplanned contact impacts, resulting in falling down or malfunctioning. Therefore, in humanoid systems, (1) balance control, (2) self-state estimation, and (3) in-body communication must have both fault mitigation and fault tolerance against contact impact. In this paper, we propose a system where robot can cope with contact impact through flexible contact by torque control, correction of estimation using contact state changes, and error tolerant communication based on control cycles. We verified the effectiveness of the proposed method through experiments of walking outdoors continuously, moving into a narrow environment, retrieving an object, and transporting an object over uneven terrain using the life-sized humanoid robot JAXON.

Humanoid motions such as walking on uneven terrain require the accurate force and position control. Joint torque control systems are suitable for position and force control, but are prone to friction and other modeling errors. In this paper, we solve this problem not with one specific layer, but rather with multiple layers that complement each other. We propose a hierarchical whole-body torque control method using four layers: friction compensation based on a vibration-suppressed model, whole-body resolved acceleration control using priority, center-of-gravity acceleration control based on foot-guided control, and landing position time modification based on capture point. We incorporate this torque control method into a system that is fault tolerant to contact impact. We verify through walking experiments that the proposed methods can control the life-sized humanoid robot driven by high-reduction ratio joints by whole-body torque control without a torque sensor or joint position control, and that it enables the robot to move and even transport an object on outdoor uneven terrain.

A wire module is coarsely attached to a moving object in the environment, such as a ceiling rail or a crane. Then, various sensor data are acquired while operating the wire module to move the environmental motion target. Focusing on the correlation between the control target and these data, the system solves an inverse problem through machine learning to achieve the environmental state desired by the user.

In order to achieve both safety and speed in cooperative manipulators, it is important to reduce the weight of the moving parts. Coupled tendon-driven manipulators can reduce the weight of moving parts by using a redundant number of wires in relation to the number of joints and integrating the motor into the root link. On the other hand, the joint structure tends to be large and heavy because a large-diameter pulley is required every time a wire passes through a joint. In this study, we propose a Passive 3D Wire Alignment System as a lightweight and low-friction mechanism that enables the transmission of wires across multiple joints. Furthermore, a lightweight 7-DoF manipulator using this mechanism is fabricated, and the usefulness of the mechanism is verified.

In order for a humanoid robot to balance on the movable ground, balance feedback control in response to its unpredictable movement is required. However, feedback control in response to ground movement has the following two issues, (A) Interaction between the ground dynamics and the balance control may cause vibration. (B) The balance control may rather deteriorate the stability due to the response delay. To solve these problems, this study proposes the support foot acceleration term in the walking stabilizer and gives its gain by considering the following two conditions, (A) Avoiding steady-state vibration in a two-mass linear inverted pendulum model on an arbitrary ground, and (B) reducing the influence of inertial forces resulting from the delay of ZMP feedback. Experiments with a life-size humanoid JAXON verified the steady-state vibration phenomenon and improved the stability of acceleration and deceleration when boarding the Two-Wheeled Scooter.

Humanoids operate in repeated contact and non-contact with their environment and so the motion of humanoids such as walking on uneven terrain or in a narrow space requires the accurate force and position control. Joint torque control systems are suitable for position and force control, but are prone to friction and other modeling errors. To solve this problem, methods have been proposed to realize torque control in combination with joint position control systems or by improving joint structures such as sensors and actuators, but these methods have problems such as response delay and increased weight and volume. Thus, it is difficult to achieve motion of life-sized humanoids by whole-body torque control. In this paper, we solve challenges not with one specific layer, but rather with multiple layers that complement each other. We propose a hierarchical whole-body torque control method using four layers: friction compensation based on a vibration-suppressed model, whole-body resolved acceleration control using priority, center-of-gravity acceleration control based on foot-guided control, and landing position time modification based on capture point. We verify through walking experiments that the proposed methods can control the life-sized humanoid robot driven by high-reduction ratio joints by whole-body torque control without a torque sensor or joint position control, and that it enables the robot to move and even transport an object on outdoor uneven terrain.

Muscle-tendon complexes that fill the body may also fill the gap between robots and humans. Several wire-driven humanoids have been developed, and their advantage is that they are soft and easily adaptable to the environment. However, wires are difficult to handle, and contact with the environment by wire parts causes malfunctions. In addition, the robot should look more human-like in order to coexist with humans. In this study, we approach these problems by proposing a driving method called wire-wound Muscle Tendon Complex (ww-MTC), which incorporates muscle expansion into the wire drive. We have realized the concept in a two-dimensional geometry and developed a 1-axis 3-muscle test robot with ww-MTC. The test robot was confirmed to suppress wire loosening, interference, and abrasion. It was also confirmed that the entire body can be in contact with the environment, and giving the robot a muscular appearance.

Quadrotor navigation for obstacle avoidance is an indispensable aspect of autonomous flight. Policy-based approaches have been extensively studied to address this problem. However, these approaches have limitations in terms of hyperparameter specificity for a given navigation system, as the learned policy can only be utilized in the same scale and dynamics as in the learning phase. To overcome this issue, we propose a transferable policy method that can be implemented across different dynamics and controller systems. This can be accomplished by (1) training a high-gain and agile quadrotor to achieve a broad range of velocities, without sacrificing performance for agile multirotor, (2) during execution, adjusting the velocity by the ratio of the maximum gain of the quadrotor to avoid high-speed collisions, and (3) randomizing only the size of the multirotor used to investigate collisions during the learning process to reduce the domain to randomize and achieve the desired policy in smaller timesteps. Our evaluation demonstrates that a universal policy, without prior knowledge of obstacles, can evade obstacles for multirotors ranging from 0.5 meters to 1.0 meter in size and across various fields. To the best of our knowledge, this is the first study to demonstrate the use of a universal policy for maneuvering tasks in various types of multirotors.

This study reports the development of a prosthesis prototype using elemental technology of tendon-driven humanoids under an industry-government-academia collaboration. Wire actuation module which is used in tendon-driven humanoids was adopted for the prosthesis. Thanks to the implementation of a joint controller of those humanoids, joints of the prosthesis can be controlled by position or torque. Two switch controllers are used as a control interface of the prosthesis. As basic experiments of the prosthesis, preliminary walk control and automobile pedal operation were conducted to demonstrate the basic performance of the prosthesis. In the pedal operation experiment, the effectiveness of torque control was confirmed based on the comparison between torque and position control. Furthermore, the experiment of automobile driving by pedal operation was successfully conducted in the road-replicated experimental field.

Human-like environment recognition by musculoskeletal humanoids is important for task realization in real complex environments and for use as dummies for test subjects. Humans integrate various sensory information to perceive their surroundings, and hearing is particularly useful for recognizing objects out of view or out of touch. In this research, we aim to realize human-like auditory environmental recognition and task realization for musculoskeletal humanoids by equipping them with a human-like auditory processing system. Humans realize sound-based environmental recognition by estimating directions of the sound sources and detecting environmental sounds based on changes in the time and frequency domain of incoming sounds and the integration of auditory information in the central nervous system. We propose a human mimetic auditory information processing system, which consists of three components: the human mimetic binaural ear unit, which mimics human ear structure and characteristics, the sound source direction estimation system, and the environmental sound detection system, which mimics processing in the central nervous system. We apply it to Musashi, a human mimetic musculoskeletal humanoid, and have it perform tasks that require sound information outside of view in real noisy environments to confirm the usefulness of the proposed methods.

Musculoskeletal humanoids have various biomimetic advantages, of which redundant muscle arrangement is one of the most important features. This feature enables variable stiffness control and allows the robot to keep moving its joints even if one of the redundant muscles breaks, but this has been rarely explored. In this study, we construct a neural network that represents the relationship among sensors in the flexible and difficult-to-modelize body of the musculoskeletal humanoid, and by learning this neural network, accurate motions can be achieved. In order to take advantage of the redundancy of muscles, we discuss the use of this network for muscle rupture detection, online update of the intersensory relationship considering the muscle rupture, and body control and state estimation using the muscle rupture information. This study explains a method of constructing a musculoskeletal humanoid that continues to move and perform tasks robustly even when one muscle breaks.

In this paper, we describe the demonstration experiments on the theme of driving an automobile with tendon-driven humanoids and the social cooperation activities under industry-government-academia collaboration. In the demonstration experiment by Kengoro conducted at Toyota City Hall, short-distance driving of the electric vehicle COMS by teleoperation was succeeded. In the demonstration experiment by Musashi conducted at Toyota Ecoful Town, in the stand-alone system setup composed of Musashi and COMS, the demonstration experiment was succeeded, in which COMS was stopped by the brake pedal operation after human recognition while the COMS drove forward. In addition, the development of prostheses is proceeding as an initiative to apply the knowledge obtained in the above demonstration experiments to the development of human assistance technology. As basic experiments of the developed prosthesis prototype, pedal operation and automobile driving with prosthesis were conducted, and a possibility of a prosthesis–vehicle cooperation was gained.

Abstract Background Physical frailty is related to adverse outcomes, and poor oral health has been linked to malnourishment. Subjective measures of oral health‐related quality of life (OHRQoL) have been used as indicators of the oral health problems of older adults, and they have been associated with malnourishment. This study aimed to assess OHRQoL's association with physical frailty. Methods Cross‐sectional study was conducted using data from the Nagasaki Islands Study that enrolled participants aged ≥60 years at Japanese national medical check‐ups from 2014 to 2019. Physical frailty phenotype criteria were determined using the modified Fried frailty phenotype model. OHRQoL was assessed using the Geriatric Oral Health Assessment Index (GOHAI). Dentists conducted clinical dental examinations. Simple correlation and linear regression analyses were performed to investigate the associations of number of physical frailty phenotype criteria with GOHAI and other oral health indicators. Results Among 1341 participants with a mean age of 72 years, GOHAI score was significantly associated with number of physical frailty phenotype criteria (B = −0.01, 95% confidence interval: −0.02 to −0.01, p < 0.001). The association remained significant after adjustment for age, sex, body mass index, history of hypertension, history of diabetes mellitus, smoking status, Kessler‐6 score, and number of remaining teeth. Conclusions Oral health‐related quality of life was associated with physical frailty in Japanese community‐dwelling older adults.

Self-healing function is a promising approach for damage management of high-load robot applications such as legged robots. Although the function is getting major in soft robotics, its application to life-sized "stiff" robots is of relatively minor interest. Although the authors have devised several self-healing tensile modules for tendon-driven robots, the design guideline to satisfy the large load endurance and large stroke is still unclear. The paper focuses on the parametric design for unleaked liquid-assisted healing of low melting point alloy structure. The method was validated with a benchtop module test. Moreover, the module enabled tendon-driven monopod testbed to perform squat motion three times after the landing impact fracture and the self-healing sequence, which was never accomplished.

A human-being often does contact-rich behaviors in daily lives. Contact-rich behaviors require human or robots to handle friction force between environment and their body. We propose contact-state transition model to achive contact-rich behaviors. The contact-state transition model represents dynamics of a robot and an environment which contacts with the robot. First, we confirm the effectiveness of contact-state transition model by conducting the hip shuffle experiment on the chair. Finally, we conbine contact-state transition model and reinforcement learning, and realize sit back motion in the musculoskeletal humanoid.

Two-dimensional multilinked structures can benefit aerial robots in both maneuvering and manipulation because of their deformation ability. However, certain types of singular forms must be avoided during deformation. Hence, an additional 1 Degrees-of-Freedom (DoF) vectorable propeller is employed in this work to overcome singular forms by properly changing the thrust direction. In this letter, we first extend modeling and control methods from our previous works for an under-actuated model whose thrust forces are not unidirectional. We then propose a planning method for the vectoring angles to solve the singularity by maximizing the controllability under arbitrary robot forms. Finally, we demonstrate the feasibility of the proposed methods by experiments where a quad-type model is used to perform trajectory tracking under challenging forms, such as a line-shape form, and the deformation passing these challenging forms.

Introduction: Reduced tongue pressure is one of the causes of dysphagia. The purpose of this study was to investigate the relationship between dental findings and tongue pressure, and whether prosthetic treatment prevents reduced tongue pressure. Methods: This was a cross-sectional study. Participants were 745 community-dwelling adults and elderly persons in the Goto Islands in Nagasaki, who underwent a health checkup for residents in 2015 and 2016. Data were collected on gender; age; grip strength; hemoglobin; Creatinine (Cr); glycated hemoglobin (HbA1c); history of stroke; smoking, drinking, exercise, and walking habits; number of teeth; wearing of removable dentures; functional units of natural teeth (n-FTUs), fixed prostheses (nif-FTUs), and removable dentures (t-FTUs); and tongue pressure. The associations between each variable and tongue pressure were examined using multiple regression analysis. Next, those with 3 or fewer n-FTUs were selected, and differences in tongue pressure were compared between those with 3 or fewer nif-FTUs and those with 4 or more nif-FTUs, using a propensity score matching method. Results: Male gender, weak grip strength, low HbA1c, no drinking, and a low number of teeth were independent factors significantly associated with lower tongue pressure. Among participants with 3 or fewer n-FTUs, the 43 with 4 or more nif-FTUs showed significantly higher tongue pressure than the 43 with 3 or fewer nif-FTUs after propensity score matching, although the number of t-FTUs was not associated with tongue pressure. Discussion/Conclusion: Tooth loss was significantly associated with lower tongue pressure. It was suggested that fixed prosthesis treatment might prevent the reduction of tongue pressure, but removable dentures did not have such an effect.

Musculoskeletal humanoids have various biomimetic advantages, and the redundant muscle arrangement allowing for variable stiffness control is one of the most important. In this study, we focus on one feature of the redundancy, which enables the humanoid to keep moving even if one of its muscles breaks, an advantage that has not been dealt with in many studies. In order to make the most of this advantage, the design of muscle arrangement is optimized by considering the maximization of minimum available torque that can be exerted when one muscle breaks. This method is applied to the elbow of a musculoskeletal humanoid Musashi with simulations, the design policy is extracted from the optimization results, and its effectiveness is confirmed with the actual robot.