Penguins utilize their whole-body degrees of freedom, including not only their wings but also their heads and hind legs, to achieve high locomotion capabilities such as rapid change of direction and acceleration. The authors’ previous behavioral observations and simulations indicate that the coordination of the wings and head achieves high-speed swimming. In this study, we aim to verify the effect of wing-head coordination on swimming speed. In this paper, we developed a penguin-like robot for real-world demonstration. Wing flapping mechanism and head flexion and extension mechanism are built into the developed robot. The robot can verify the effect of the coordination between the wings and the head on swimming speed by arbitrarily changing the phase difference between head flexion-extension motion and wing flapping motion.

We focused on the motion of catching a fly ball in the outfield. In order to develop a training system using voice instructions, we examined the effects of voice instructions on the catching behavior of the fly ball and the most appropriate voice instructions among several voice instructions. The voice instructions were effective in catching the fly ball. It was found that the most effective instruction for experienced players was the instruction with front and back information, while the most effective instruction for inexperienced players was the instruction with a large amount of information. It is considered that changing the method of voice instructions according to the level of experience will lead to effective training. We propose a method of wearing VR with less burden on the neck and a safe method of using VR to understand the surrounding environment while using VR.

In badminton, beginners are considered to be at high risk of increasing elbow injuries. It is also said that when the shoulder abduction angle is close to 90[deg], the large muscle groups of the trunk and shoulder periphery can exert greater tension and energy and swing the racket faster, but the impact of swinging the racket faster increases the strain on the elbow and shoulder joints, increasing the risk of injury incidence. Considering these stories, this study investigated the differences in each joint angle by measuring the high clear operation of skilled players and beginners who have or may have elbow pain. The results suggest that external and internal rotation of the shoulder joint is related to elbow pain.

In recent years, many androids have appeared on the market and behave like humans, but people sometimes feel uncomfortable when they see them move. This may be due to the effect of the uncanny valley, in which the closer a robot is to a human, the less familiar it becomes. Also, when comparing the movements of two androids, one android may feel unnatural. This study examined the difference in unnaturalness between two androids by analyzing the relationship between head and chest movements through motion analysis of a human and an android.

This paper reports the development of an elbow joint mechanism and its potential as a safe and versatile joint mechanism. Humans possess superior abilities such as dexterity and flexibility. We propose a new elbow joint mechanism that mimics ligaments using a three-dimensional model of the skeleton and an elastic belt made of polyurethane in order to mimic these superior abilities. First, the structure of the human elbow joint is described in order to mimic the ligament structure. Next, the elbow joint mechanism developed in accordance with the human ligament arrangement is introduced and its operation is confirmed by flexion movements. Finally, as a performance evaluation experiment using the developed elbow joint mechanism, the special characteristic are investigated by tension and compression tests.

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 communication robot with a human-like touch is useful. Warmth, softness, and the capability to get in motion are important to realize such a robot. We proposed the use of polymer materials and drive using a water-cooled engine for the cogeneration of body heat to conserve energy. The soft flesh made from gel and joints made of polymer, a circulation system to heat the flesh with coolant, and multiple clutches to move the robot‘s arm with a single driving source were implemented partly. For the fingers, we proposed a braid system to reduce the number of degrees of freedom for driving. In tests with an induction motor and sheath heater, driving the robot‘s arm and fingers by a single driving source and warming of the arm body surface with warm water were achieved. The design of the engine unit and the thermal drive of the polymer actuator were studied.

In this study, we measured the GRF patterns during walking on several ground conditions using shoes equipped with MEMS 6-axis force sensors in their insoles. The MEMS 6-axis force sensors were placed at the thenar, hypothenar, and heel, and they measured 3-axis force and momentum with a measurement rate of 50 Hz. Measurements were conducted on level walking, sand walking, and slope ascent and descent. The data were analyzed for the three axial forces on each measurement. In the sand walking measurement, the GRF for each axis was smaller and had high variability compared to level walking. The slope ascent and descent measurements showed that the hypothenar produced a greater propulsive force when ascending, and the heel produced a braking force when descending.

In this research, we developed a humanoid robot with suction mechanisms to realize footwork in badminton. First, the investigation of the suction force of suckers showed that the stability of the robot had improved. By improving the stability, the robot was able to perform a large and quick-stepping motion that was not possible with existing robots. The width of the step was about 0.9m, and the horizontal velocity of the hip of the robot was about 1.1m/s. Compared to the human stepping motion, the speed of the robot’s stepping motion was inferior, but the width of the step was no inferior. In addition, the estimated ZMP during the stepping motion showed the effectiveness of suction mechanisms in the realization of badminton footwork.

Recent studies have reported that human runners may run more efficiently with passive equipment. In this study, we examined whether a passive external tendon connecting the legs would improve running performance and running economy. Our results obtained from energy expenditure analysis showed that the passive external tendon connecting the legs had a positive impact on running economy, supporting previous studies. On the other hand, the results also showed that the external tendon had a negative impact in some cases. For a runner to obtain effective performance with the passive external tendon, not only the strength of the external tendon but also the runner's running style and posture must be fully considered. Further research including deeper kinematic and kinetic perspectives is needed to find the optimal "sweet spot" of intervention that can fully exploit the mechanical function of the passive external tendon.

Archery is a sport in which it is important to repeat the same shooting form and to improve its repeatability. For this reason, it is essential to understand and master the basic shooting form. However, there is a shortage of instructors, and in university club activities, the burden of teaching is placed on the senior students. A possible solution to this problem is the use of a practice support system based on sports video analysis, but existing systems have a fixed positional relationship between the camera and the person to be measured, and it is necessary to create the same environment each time the system is used. Therefore, this study proposes a sports practice support system that uses 3D skeletal point information extracted from video images to evaluate the learner’s shooting style in terms of points, eliminating the need to fix the positional relationship between the camera and the learner and allowing easy recognition of points that need improvement.

UAVs have challenges in terms of safety and cruising range. A flapping flying robot has the potential to be a superior UAV in these aspects. The flapping motion is an oscillating motion and is safe. In addition, flapping flight robots don’t necessarily require flapping motion to sustain flight. In this study, we focused on bounding flight and aimed to achieve it with a flapping flight robot. Bounding flight is a flight method that many small birds perform by intermittently folding their wings during flapping flight. Therefore, to perform bounding flight, it is necessary to perform both the flapping motion and the wing folding motion. We have developed a flapping wing flying robot that can perform these motions and mimic bounding flight.

In recent years, the development of aircrafts with good environmental performance is required to address the global warming problem, and the reduction of aerodynamic drag is an important issue in the development of aircrafts. A wingtip vortex suppression device called a winglet is mounted to the tip of airplane wings. Winglets are modeled after the upturned feathers on the wingtips of gliding hawks and eagles. Winglets have been shown to improve the performance of airplanes by 10% or more. The development of the shape of winglets must be considered in combination with other aspects such as the increase in weight and drag due to the addition of winglets. In this study, we developed flapping wing aircrafts with winglets and compared performances in order to find the optimal winglet mounting angle for flight.

In this study, falling object which was inspired by a lauan seed was designed. The falling object's posture and falling speed are stabilized by the auto rotational wing. Also, the non-rotational part controls the scattering area by depending on its part angle. It was confirmed that the non-rotational part expanded the scattering area by experiments.

In this paper, we propose a composite modeling method that combines high-viscosity cure inhibitors, high-viscosity lubricants, modeling of adhesive surfaces, fiber reinforcement, and other techniques. This method allows for customization of molding characteristics to some extent by utilizing different methods, such as adding inhibitors and lubricants, incorporating fabrics, and molding surface textures. The degree of lubrication, adhesion, and tearing strength can be achieved using these techniques. In the latter half of the presentation, we demonstrate a robot finger modeled using this method, which is capable of smooth tendon-driven motion.

Continuum robots are suitable structures for operations in difficult-to-reach environments, such as narrow environments or with many obstacles. Holding the shape and changing the stiffness of their body is effective for applying the robot to various situations. However, conventional methods had difficulty to construct high shape-holding performance, thin body, elastic properties, and large bending range. This research proposes a wire binding mechanism called "String Binder" that can firmly hold the shape of the continuum body. The robot mainly consists of a flexible tube, longitudinal wire, and circumferential wires. When the tube is pressurized, the circumferential wire strongly ties the longitudinal wire and locks its position. This paper presents the basic principles of the mechanism and verifies the validity of the concept.

In the micromouse competition, small autonomous mobile robots run at high speed in mazes. The robot turns a corner at a steep angle in this competition. However, the recent competition is marked by speeding up. There is a limit to driving performance with motor current control braking. Electro-rheological (ER) fluid actuators have the potential of soft brake function for the small autonomous mobile robot because the ER fluid can easy and quickly control changes in viscosity by changing the electric field. In this paper the double cylindrical type of mechanical brake device using the ER fluid for the small robot was designed and fabricated. The performance of the ER braking device was verified through experiments.

This paper presents a driver for unconstrained poppet valves based on a CR elecronic oscillator. Unconstrained poppet valves are small-sized and lightweight, capable of controlling air flow around 0.5 MPa up to 10 L/min. Drivers for unconstrained poppet valves require generation of pulse signals and amplification of driving current. We applied CR elecronic oscillator circuit to realize these requirements on a driver. We found that the developed chopper-based driver could control unconstrained poppet valves

McKibben pneumatic artificial muscle (MPA) is a kind of artificial muscle used in musculoskeletal robots. This actuator contracts and exerts tension like a muscle by applying compressed air. In musculoskeletal models with MPAs, the MPAs are antagonistically placed around joints, and the joints are driven by antagonistic drives, such as in real animals. Therefore, the antagonistic MPAs should cooperate with each other to generate various movements. However, in previous studies, the cooperative patterns for dynamic movements such as walking or jumping were typically found by trial and error. As a method to generate the cooperative patterns autonomously, we have proposed a tension feedback cooperative control law. In our previous study, we applied the control law to a legged model with four muscles (two pairs of antagonist’s muscles) around one joint. We confirmed the autonomous coordination of antagonistic muscle groups through simulations. In this paper, the legged robot was actually developed, and the validity of the simulation results was verified through experiments on the robot. The experimental results confirm that the control law changes the air inputs to the MPAs autonomously and that the coordination of the antagonist muscle groups and the periodic motion are generated autonomously.

Multi-fingered hands can achieve stable grasping manipulations as their fingers synchronously make contacts with an object and demonstrate skillful motions such as dragging of unknown objects. Abundant tactile is useful specifically for multi-fingered manipulation which causes visual occlusions. However, generating dexterous motions with high density tactile sensors is difficult due to complicated contact states. In this paper, we propose a novel deep predictive learning approach using Vision-Transformer (ViT) and Long-Short Term Memory (LSTM). In this method the ViT extracts tactile information that is important for motions, while the LSTM remembers the orientation and characteristics of objects from tactile information and link them to motions. The model uses the 16 joint angle measurements and 912 tactile measurements which are distributed on the finger and the palm of the Allegro Hand. The model achieved high success rates of 95% for dragging motion by adapting to the target object orientation and feature.

In food factories, serving is still done manually by people in many cases. To solve the problem of quantitative serving, which is one of the challenges in robot operation, we developed a Chobit Hand that can precisely grasp only one strand of spaghetti. In this paper, we simulated whether installing the Chobit Hand on a multihead weigher would increase the efficiency of the weigher. As a result, the number of combinations could be increased by more than 1 set by replacing some of the existing hands with the Chobit Hand. It was also found that the introduction of the Chobit Hand reduces the variation in food product weight, allowing the producer to reduce the cost of food.

A three-fingered hand grasping system for flexible objects is constructed, and its validity is confirmed by fundamental experiments. A 3-fingered hand is attached to the tip of a 6-DOF manipulator, and a hamburger as a flexible object is grasped based on images obtained from a 3D camera. The optimum three fingertip positions and the grasping forces are calculated that minimizes the pressing force of the fingers under the condition that the hamburger is not dropped. As a result, it is found that the optimal grasping position becomes one of two patterns depending on the hamburger shape. However, one of them has a high probability of dropping objects due to finger position errors, which indicates that an additional evaluation function is required to prevent dropping in practical use.

In the food and other manufacturing industries, where the introduction of robots is expected to be promoted, hygienic and stable grasping technology is essential because of the need to handle goods with diverse characteristics. However, for companies, the high market cost of robot hands that meet the requirements is one of the factors restraining their introduction. In order to solve the problem of hygiene by using disposable robot hand, we devised an inexpensive disposable robot hand made of paper, verified the relationship between materials, structure, and grasping force, and studied the primary modeling of the force transmission mechanism of the robot hand in order to reduce the size and increase the grasping force.

In this study，a mechanism for a gripper that uses a wire-driven two-degree-of-freedom manipulator to perform grasping by suction at the suction part attached to the tip，without using image processing is developed and evaluated. To achieve this，a torque equation for each joint to align the suction part with the grasped object is derived. To satisfy this equation，non-circular pulleys are attached to each joint. We conducted experiments on the grasping and suction operations of a suction-type gripper equipped with the necessary functions for the gripper. Finally，we demonstrated that it operates well in many cases within its range of motion.

When grasping a fragile object, small grasping force and gripper compliance are required to avoid deformation and damage of the object. In this study, linear motor-based mechanism is used to achieve small gripping force and compliance without using force sensors. The linear motor mechanism has low friction and no reduction gear, so its back drivability is high. The robot hand consists of two linear motor mechanisms, a guide rail, and a frame. The thrust force can be controlled through a driver. By overlapping the strokes of the two mechanisms, the redundancy of the opening and closing motions can be achieved. Due to this redundancy, it is possible to adjust the position after grasping. Experiments were conducted on a hand prototype and small thrust force was confirmed. In addition, we attached this robot hand to a robot arm and conducted a grasping experiment using fragile object.

Myriapods exhibit various locomotor patterns depending on their species or situation while walking. Some species walk by direct wave gait, whereas other species walk by retrograde wave gait. Furthermore, some species employ body undulation as well during retrograde wave gait for fast walking. Our research group hypothesizes that there is a common control principle underlying these body–limb coordination patterns. Our previous study using two-dimensional simulations showed that they could be well reproduced by a single control principle that describes a simple coordination rule between the legs and body. In this study, we developed a myriapod-like robot to demonstrate that the control principle is valid even under three-dimensional body motion in the real world.

The synchronization phenomenon refers to the gradual synchronization of simple elements with different rhythms due to mutual influence and is observed across various fields, e.g., natural sciences to humanities and social sciences. Synchronization can be observed at various levels in the human brain, for example, epilepsy due to the abnormal synchronous firing of neurons, resulting in functional behaviors. Predicting and analyzing synchronization dynamics could be important for inducing or preventing functional behaviors of synchronization. However, most studies on synchronization have focused on complex oscillator systems, and few have predicted and analyzed the synchronization of basic pendulums using machine learning. In this study, we reproduce the synchronization of pendulums in a physics simulation with an easy condition setting and use machine learning to predict how the dynamics change under various conditions. We investigate which machine learning model is best for predicting synchronization phenomena.

This paper presents safe coverage control algorithms for multi-agent systems with time-varying uncertainties. This covers the multi-agent control problems like coverage control with obstacle avoidance. The coverage control problem is concerned with allocating agents within an area of interest to optimize a coverage metric. An optimized allocation is achieved through centroidal Voronoi tesselation (CVT) coverage controller is constructed to realize the CVT in the presence of time-varying disturbances. The obstacle avoidance problem requires a decentralized control strategy that guarantees collision-free maneuvers in an unknown environment. A control barrier function (CBF) based controller is developed to ensure collision avoidance taking into account the actuator fault, which would seriously affect the system performance. To verify the validity of the proposed controllers, simulator has been constructed under the python environment. The CBF based controllers have been verified under simulations.

Animals can change their behavior adaptively in diverse environments. In this study, we developed a small quadruped robot that can reproduce the reflexive behavior of animals to investigate why quadruped animals exhibit specific gaits. We construct reflex circuits based on the results of previous experiments on walking cats. In walking experiments, the robot exhibited pace and trot gaits, depending on parameters about a motor angle. The contribution of this study is showing that the quadruped robot can exhibit two gaits only with reflex control of each leg. These results suggest that leg length parameters may change the gait pattern of animals.

In recent years, it is required that autonomous construction machines (backhoe and dump truck) work together in an unlimited (unknown/unexpected/uncertain) environment to realize collaborative sediment transportation. In an unlimited environment, anomalies such as stuck or malfunctioning of autonomous construction machines occur, so a group of autonomous construction machines requires dynamic team organization. However, such a system has not been established. Thus, this study designs the autonomous flow to dynamically reorganize teams based on an index called the performance to evaluate sediment transport efficiency. Using Vortex Studio Simulation, this paper confirms that teams can be dynamically organized to keep team performance adequate even when an anomaly occurs in which the construction machine gets stuck for a one-team sand and soil transportation.

In the case of a swarm robot system in an unlimited environment such as a disaster site, it is necessary to take an approach in which robots autonomously solve each other’s anomalies such as stuck robots or malfunctions. This paper proposes an example of an autonomous anomaly diagnosis and rescue architecture with swarm robots. Through verification using actual robots equipped with rescue mechanisms, this paper presents an example of architecture design for responding to an anomaly.

In this paper, the grasping performance was evaluated for a dual-arm robot arm with a jamming gripper attached to the tip of the arm. The developed robot has dual arms with 3 degrees of freedom attached to the front of the opposing two-wheeled body. By deforming the jamming gripper according to the shape of the object, it is possible to grip objects with curved surfaces such as cylinders and discs that are normally difficult to grasp. From the experimental results, it was found that the grasping force is affected by the shape of the object and the gripping position. The maximum grasping force was 39.4N in the X-axis direction and 43.3N in the Z-axis direction.

In this paper, we propose an accurate and fast method checking collision of ellipses. Ellipse collision detection techniques are used in various applications such as robot collision avoidance and material packing. Our proposed collision detection method is based on the Minkowski sum, which is the union of two ellipses as a subset. Ours approximates the Minkowski sum of ellipses in the form of a superellipse. We evaluate the accuracy and computation time of the Minkowski sum for various combinations of ellipses and describe the simplicity of the collision determination.

This paper proposes the novel robotic gripper that realizes the three functional operations: grasping object, rotating object, and pull-in object. The developed gripper achieves these functions by a single motor. The main target task of the developed gripper is grasping objects in narrow spaces, such as the picking up object from a box in a shelf. To achieves three functions, the gripper adopted two self-motion switching mechanisms. One is for switching the gripper behavior automatically between the grasping and rotating object modes according to the fingertip posture. Another switches the gripper behavior between the grasping and pull-in modes according to the grasping force applied to the fingertip. This study proposes the structure of the griper and evaluated the gripper experimentally.

In this study, we propose a novel end-effector for picking and sorting operations of various fabric parts placed on a cutting machine in sewing factories. We also propose a method for grasping a whole body of the fabric parts by folding back and forth a long fabric into a smaller size, which facilitates packing and sorting of each part of the fabric and reduces the requirement for the height movement range of the robot.

There are many examples of research, development, and practical application of robotic hands. There are various types of hands, such as hands for industrial products, hands for food, and hands for specific food. There are few examples of research on hands for grasping composite objects including rigid and flexible bodies. Therefore, this research aims to develop a hand that can grasp a composite object that including rigid and flexible bodies. We analyzed the grasping motion of a worker from the actual dango manufacturing process and defined the design requirements for the hand development. Based on the design requirements, a prototype hand was designed and fabricated.

This paper describes the hysteresis characteristics of the double-acting artificial muscles we are developing. General pneumatic artificial muscles have hysteresis characteristics, which are difficult to control and hinder the widespread use of artificial muscles. Therefore, we developed a double-acting type artificial muscle that aims to eliminate hysteresis by controlling two systems of pressure. Feedback control was applied to eliminate hysteresis under various operating conditions. The possibility of improving hysteresis characteristics by this method was confirmed through analysis and experiments, and was found to be effective.

A McKibben-type pneumatic actuator (MPA) is a soft actuator that exerts tension by applying compressed air to expand a rubber tube. Although electro-pneumatic regulators are usually used to control air pressure, most of them are large and expensive. This study utilizes a dynamic quantizer to control the MPA with a small solenoid valve that can only open and close the valve instead of an electro-pneumatic regulator. A dynamic quantizer is one of the quantizers that converts continuous signals to discrete signals. Our previous study confirmed that tension control of MPA under isometric conditions could be realized using a dynamic quantizer. This paper presents the results of implementing a dynamic quantizer to control the length of the MPA.

This paper discusses both a structure and a system for a soft suction cup for the purpose of sucking and grasping organs. Recently, in laparoscopic surgery, there is a demand for a means for easily and safely suppressing an organ covering over an affected part. The authors propose a flexible suction cup, which is designed in order to suck and grasp deformable organs, by utilizing the negative pressure generated by quasi-indirectly sucking method by the deformation of the elastic membrane and by increasing the contact area with the ring-shaped groove on the surface of the sucker.

Rapid and precise understanding of 3D objects in virtual reality is crucial for proficient manipulation. Generally, relying solely on a force feedback device falls short in conveying intricate shapes, such as the edges of 3D objects, and it is deemed necessary to supplement it with appropriate cutaneous sensory inputs. Electro-tactile stimulation has the potential to provide high-resolution cutaneous sensory inputs when incorporated with a force feedback device. We devised a system that presents cutaneous inputs along an object’s edge through electrical stimulation and reactive force through a force feedback device. We evaluated its impact on 3D shape perception under three scenarios: force feedback alone, cutaneous feedback alone, and combined sensory presentation. Results from experiments on identifying four types of column shapes in single-finger contact and two-fingers grasping indicate that combined presentation of force and electro-tactile sensation significantly hastens shape differentiation time and facilitates more efficient recognition of 3D objects.

The Thermal Grill Illusion (TGI) is a phenomenon that causes a burning sensation or pain when warm and cold stimuli alternate. The chemical-based TGI method can provide a prolonged painful experience with low energy. However, it has the challenge of requiring a time lag between the application of two chemicals to produce a strong TGI. To overcome this limitation, we investigated the impact of the placement of chemical stimuli on TGI perception. In our first study, we divided the stimulus area into two parts and compared simple placements like vertical and horizontal orientations. Our findings show that TGI perception is influenced by the placement of stimuli, and the strongest TGI is produced when capsaicin is applied proximally on the forearm and menthol distally. In our second study, we compared a more complex placement, a grid-like stimulus placement, to the strongest placement from the first study.

Although grip strength, tactile softness, and temperature are considered to be important cues of "human skin-like feeling" in situations where human skin comes into contact with each other, not many studies on the reproduction of human skin sensation have focused on the temperature in the past. In this study, we focused on skin temperature as one element of human skin feeling and measured skin surface temperature during human skin contact. The temperatures were measured by making contact with fingertips, palms, backs of hands, and forearms of other subjects with index finger, and palm with palm. As a result of the experiment, differences in skin surface temperature changes were observed for each site.

Mechanical vibratory stimuli to the thoraco-abdominal during watching movies magnifies emotional experiences. We hypothesized that the vibration to thoraco abdominal stimulates the vagus nerves, which is distributed to the organs of the upper body, and that magnifies the emotional experience. To investigate this hypothesis, we compared the effects of vibratory stimuli on fear between the thoraco abdominal and the fingers which have the highest tactile sensitivity in the body during watching horror movies. The effects of the vibratory stimuli to the thoraco abdominal on subjective evaluation and skin conductance response amplitudes were greater than those to the finger. These results suggest that mechanical vibratory stimuli to the vagus nerves are the primary factor in the magnification of the emotional experience.

Phantom Sensation (PhS) is a haptic illusion discovered by Georg von Békésy, also known as funneling illusion. PhS generates tactile sensation between more than one tactile stimulus when they are presented at the same time on the skin. In previous studies, funneling illusion was produced by the same kind of stimuli (e.g., mechanical, or electrical stimuli). While on the other hand, this study aims to investigate the nature of funneling illusion with different types of tactile stimuli. We built the experimental system and conducted evaluation experiment. This system could produce different kinds of stimuli to two points on subject's skin at the same time. This system will provide new knowledge of the nature of the illusion. An important question to consider is whether painful stimuli also make the illusional tactile stimulus painful and whether painful stimuli affect the size of the illusional stimulus.

This paper reports on direction discrimination thresholds at three areas on a palm. The stimuli used to measure the thresholds were pressure stimuli provided using a rectangular-shaped indentor. Previous studies have shown that pressure stimulation can invoke pseudo-force sensations at various areas on a palm, and devices that can present such stimuli have been developed. Understanding the threshold is important for the applications of such devices. Although two-point threshold and point localization of a palm have been investigated using point stimuli in the literature, discrimination ability for rectangular-shaped stimuli has not been clarified. Measurements of discrimination ability revealed that the direction discrimination thresholds were approximately 6.1 to 9.2 degrees, although they differed for each of the three palm regions.

Collaborative robots can work collaboratively with humans in the same place, they are expected to save space and work more efficiently. Proximity and tactile sensors play an important role in ensuring the safety of robots. In this study, we propose a string-like self-capacitive proximity and tactile sensor. The sensor consists of two measurement electrodes (E1 and E2), an elastic body, and a GND electrode. The sensor can be bent in one direction by making slits in electrodes that are non-stretchable metal cloths (E1 and GND electrode). The string like sensor may be wrapped around a robot of various shapes and easily mounted. The string-like self-capacitance proximity and tactile sensor can detect objects at close proximity area and detect contact conditions on contact.

Both cutaneous and kinesthetic sensations contribute to human dexterous manipulation performance. To realize skillful manipulation in VR environments, tactile and force senses must be reproduced with high realism. This study aims to develop a display in which tactile and force sensations can be presented independently by integrating a suction tactile display and a parallel-linked force display, which has low inertia, high rigidity, and high-speed operation. Design guidelines for the integration are reported, and a prototype of a suction tactile display for the integration is shown. A contact simulator with a deformable finger is also introduced for future cutaneous sensation presentation.

Stereo vibration is an innovative vibrotactile technology that extends the conventional phantom sensation to the surrounding space, representing a virtual vibration source in the external environment. This vibrotactile localization technology is expected to be combined with stereophonic sound to reinforce each other’s localization performance through multimodal perception and to be applied to enhance the reality of virtual reality. This paper reports the analysis of the interaction between stereophonic sound and stereo vibration in localization based on the results of an experiment to identify the travel paths of a virtual vibration source represented by a bracelet-type device and a virtual sound source represented by an earphone.

In recent years, massage is expected to be used to address the problems associated with the rapid increase in the elderly population in Japan. In order to realize better massage, we have been developing a massage robot with a multi-fingered hand that can respond to the physique of the person being massaged. The robot must be equipped with highly durable, compact, and multi-directional force-measuring sensors at its fingertips to ensure safe operation. We are developing a vibration-based force sensor that meets these characteristics. In this paper, we report on the relationship between the resonance frequency and the load in two directions on the prototype force sensor, and the verification of the estimation method of the load in two directions.

Marker-based estimation is one of the most commonly used methods for estimating the position and orientation of objects. However, there are many problems with markers, and marker-less estimation methods are currently being sought. A common method is to use machine learning to estimate position and posture from distance information and RGB images. However, this method has low accuracy for simple shapes. In this study, we propose a method to simplify position and pose estimation by using persistent homology, a topological theory, to extract geometrical features from distance information and predict the ground plane by machine learning.