Anthropomorphic robots will gradually become an essential part of our daily lives. The development of reliable control algorithms for them is one of the actual problems of robotics. Our research concentrates on effective locomotion algorithms development for Russian humanoid robot AR-601M. AR-601M is modelled with 41 Degrees of Freedom (DoF) in Simulink environment. We use 6 DoF in each robot leg with Virtual Height Inverted Pendulum Model and Preview control methods in order to achieve a human-like dynamically stable locomotion.

Natural walking patterns are not always based on the well-known ZMP (Zero Moment Point) criterion that has been usually applied to walking robots. This paper focuses on classifying the stable state of a foot-stepping robot. First, a parallel typed foot-stepping robot is introduced to investigate the state transition of stepping patterns. Next, the dynamical model of the foot-stepping robot is formulated, which consists of ten kinds of state including a floating state in air, in order to simulate the stepping motion. Then the experimental results are compared to the simulation results, and it is confirmed that there are two kinds of stable states that consist of usual ZMP motion and parallel motion.

Recently, legged robots have been developed to work in a dangerous space. This paper introduces a Limb-type robot which is developed to move in frame structures. However, in order to adapt the robot to works, flexibility for replacing the components of the system is necessary. This paper explains the system which integrates components such as a motion generator, sensor, configuration of the Limb-type robot and so on. “Robot Operating System(ROS)” is adopted to construct the system.

In this study, we propose dead reckoning method using an optical sensor, an azimuth sensor and a gyro sensor. By combining several sensors, even if slipping wheel is generated indoors where radio waves of the GPS do not reach, the robot performs dead reckoning that allows the self-position estimation with high precision.

This paper describes an autonomous outdoor navigation method based on view-based localization. We develop a new view-based localization method by extending Seq-SLAM which exhibits a robust place recognition performance across seasons. In mobile robot navigation, traversable region detection, as well as localization, is important. We therefore combines a view-based localization with a traversable region detection using 3D range sensors such that subgoals towards the destination are continuously chosen inside the local traversable region. We developed an autonomous navigation system by implementing the proposed method on a mobile robot, and demonstrated the validity of the method by carrying out autonomous navigation experiments in own campus.

We present a compact measurement module for estimating omnidirectional motion of a mobile robot. The module consists of an optical device with a laser to detect translation and a gyroscope to detect rotation. Laser speckle patterns are generated on a passage plane and tracked in the optical device. The translation estimated by the device is constant even in different heights from the passage plane. The zero drift in the gyroscope is automatically calibrated in case that motion is not detected by the optical device. We investigate the accuracy of the estimated translation in different internal parameters for the device. Additionally, we show estimated trajectories of a cart moving on a carpet.

A LIDAR is commonly used for autonomous mobile navigation. It detects raindrops under rainy situations since its output laser beams collide the drops. This influences against performance of the navigation, for example obstacle avoidance. In this study, we first investigate influence of rain against the LIDAR and then propose a coping method, which contains two ways, for reducing the rain's influence. The first way is a hardware-based coping method that is to add additional waterproof parts. The second way is a software-based coping method that is to remove raindrops from the LIDAR readings like a filtering algorithm. By using these methods, number of raindrop detection is significantly reduced. As a result, autonomous navigation can be performed in rainy situations and its performance is verified in a man-made environment.

This paper explains gait implementation and leg-wheel transition of leg wheel hybrid locomotion to the four legged robot named “Road Strider”. Road Strider is equipped with a passive wheel to each foot. It moves on a flat plane by wheels and on the rough terrain with legs. The Road Strider has unique joint mechanism which uses singular point to reduce the energy consumption when the robot stands. High energy-effectiveness is suitable for the various application, especially the robots acts in the extreme situation like a rescue mission. The implementation of the various gaits such as crawl, trot and roller-skating to Road Strider are explained and a new gait to transit between roller-skating and walking is proposed.

Many robots have been developed for the purpose of exploring. However, carrying robots to the disaster sites is dangerous. If it can be thrown into the disaster sites, operators will be safe. The purpose of this study is the development of durable robots. Thus, we developed quadruped robot with spherical shell. We have added a jumping function to climb over steps to it and Calculation of it by air cylinder for the realization. We found conditions for jumping of legs for jumping. We confirmed jumping preparation posture by using the inverse kinematics. Also we confirmed conditions jumping and formula of jumping directions.

In this study, to improve performance of autonomous odor source search robot, we remarked artificial olfactory system which has smell intake function similar to animal 's sniffing. The practical use of the robot is expected in various fields such as rescue and the poisonous gas counterterrorism. The olfactory system can exchange the intake speed, placement of gas sensors and shape of itself so that we can inspect a variety of intake strategies. To evaluate performance of intake strategies, we carried out odor source searching experiment by using the real robot in plural air flow environments. Under most environment, robot using air intake reached to odor source with higher probability than not using intake. On the other hand, higher intake speed does not achieved higher performance.

This paper describes the direct perception for grasping and the positioning of a robot based on the information of a depth sensor. Many conventional grasping methods have required the accurate physical parameters of an unknown object. The direct perception is the method based on the ease of grasping not using the accurate physical parameters. The grasping sense is described by inertia tensor and fuzzy inference. Through grasping experiments, the effectiveness of the direct perception method is confirmed.

In this paper, we describe an obstacle detection system using Kinect V2. This method is using point clouds providing from the Kinect V2. As a reason to use the Kinect V2, it is very cheap compared to 3D LIDARs. Moreover, I can detect 3D shapes of obstacles with high sampling rates. The point clouds are provided by the Kinect, those plane regions are detected using the LMeDs method. Removing that from all point cloud, we extract point clouds with obstacles information. The experimental results show that robots can detect obstacles, because of extracting point clouds with obstacles information. Our proposed system can detect more than height of 3.0 [cm] obstacles in the outdoor environment.

The maintenance and management of mountain forest needs special knowledge and skill, and contains various dangerous tasks. Thus, the shortage of workers causes the problem of increasing unmanaged forests. In this research, we are developing “Teleoperated mountain forest maintenance robot system”, which assist human work semiautonomously. In this paper, we will discuss and propose a performance improvement method in recognizing process of the surrounding objects in the mountain forest environment. The experimental results illustrate the validity of the proposed method.

This paper presents a cooperative moving-object tracking with multiple mobile sensor nodes (mobile robots) using in-vehicle multilayer laser scanner. Each sensor node detects moving objects such as people, cars, and motorcycles using its own multilayer laser scanner by an occupancy-grid-based method and then estimate their sizes and postures. After that, each sensor node sends the estimates to a central server. The central server combines the estimates received from mobile sensor nodes to improve the tracking accuracy, and their estimates are fed back to the mobile sensor nodes. Experimental results of tracking vehicles by two mobile sensor nodes show the performance of the proposed method.

Autonomous mobile robots have been an active research recently. Although laser range finder based navigation is known to work well in limited environment, expanding 3D geometric maps used in those methods to worldwide requires great effort. The same is true of maintaining them up-to-date. For this reason, we develop a visual navigation method using a topological map to reduce prior information and enhance scalability of the map. Topological maps only contain information about impressive places (i.e. intersections) and their linkage with roads. In the topological approach, the robot conducts three primal tasks: 1) road following between impressive places, 2) recognizing impressive places, and 3) taking action at impressive places (e. g., road selection). This paper proposes a visual place recognition method based on GIST descriptor and Support Vector Machine. The proposed method is evaluated at 13 point in outdoor walkways under several situations. As the result, the proposed method could identify all of 13 recognition points.

This paper describes a method which make sharing search and rescue (SAR) dog's investigation activities. The search and rescue activity will be operated several organizations. Therefore, it is necessary for correct decision making to share SAR dog's investigation. The authors propose a sharing method using widespread high speed communication and interface which can be operated by several operating systems. The proposed methods were evaluated by experiment. In several computers, the sharing and presentation methods were operated simultaneously. From the experiment, there were delays of displaying position data and movie data from camera. Evaluating those delays, delay of positon data is 1.5±0.85 seconds and delay of movie data is 20±6.9 seconds.

This study proposes an advanced holding force generating mechanism to improve performance of in-pipe locomotion mechanism. A high-speed locomotion mechanism using pneumatic hollow-shaft actuators, which is previously proposed by authors, has two challenges: lack of a holding force and inability to pass through a bend pipe. The proposed mechanism overcomes the challenges by applying new holding force generation method to the locomotion mechanism. First, design of previous locomotion mechanism is presented, and then challenges for further improvement are organized. Second, a design concept of a novel mechanism using self-locking phenomenon is proposed. Finally, characteristics of the holding force and passing performance through the bend pipe are investigated through experiments. Experimental results show that a retractable mechanism of pins can generate maximum holding force of 69.7 N, which is 5.2 times higher than the previous one, and improves passing performance through the bending pipe drastically.

Tethered robots often face the entangling of the cable with obstacles in uncertain disaster environments. We have proposed a spherical tether handling device that unfastens a robot's tether during surveys by releasing the tether and carrying it aside. The device is able to move forward by driving the outer shells. When the device climbs obstacles, it alters the driving power from driving the shells to pull the tether to obtain high surmountability because of a coupled differential mechanism. Specific characteristics of the coupled differential mechanism were evaluated and experiments of basic travel motion of the device were performed in this paper. The results of the experiments show that the device can climb a height of 90.9% of its diameter. We also demonstrate a scenario to handle the tether and untangle multiple tangles in an environment with several obstacles.

There were many problems with so far suction component. For example, there is a problem such as weak to force applied to the horizontal direction of the suction surface, the suction force that it is uneven surface is significantly reduced, and Low durability. We developed the suction component that overcomes the problems, and investigated the adaptability. In this paper, we report to give a prospect of creating design method of high durability suction component that can be stably adsorbed to a variety of surface condition.

Teleoperation system using past image records (SPIR) is a system that shows the image of the robot and the surrounding environment from a third person viewpoint to an operator. The image is created by superimposing a CG model reflecting a current pose of the robot on the background image captured by the camera mounted on the robot in the past. In this paper, we propose a new teleoperation system based on SPIR for a teleoperated legged robot. Legged robots are capable of moving on uneven terrain and stepping over obstacles with their legs. By using the proposed system, the operator can easily recognize legs' state of the legged robot and the surrounding environment, and can perform tasks well.

Robot technology that can work robustly even in unknown, time-varying extreme disaster environments is demanded. As such a tough robot technology, we focus on the unmanned aerial vehicle (UAV) that can perform information collection and work from the air. It is possible to efficiently grasp disaster circumstances by comparing acquired data from UAV with the terrain data acquired beforehand. Although the acquisition of the terrain data has become much easier by the development of laser scanners, dense point-clouds of a city, a prefecture or a region cannot be dealt with on a computer because of its limited memory size. In this paper, we propose a method to generate a multi-resolution database for large-scale terrain data by using quadtree-based multi-layers and compare image data acquired by a UAV with the database to acquire damage states suffering from the disaster by an earthquake.

This paper describes material identification suitable for objects in rubble. We focus on the difference of surface appearance for the material identification. We applied texture-based features: filter banks, LBP, and edge histograms. In proof experiments, we used 10 kinds of objects selected in reference to building materials, interior decoration, and objects in the indoor. The results for 7 of 10 kinds of objects showed high success rate.

This paper presents a crawling direction control for four-limbed robot on rough terrain. Crawling is useful from the viewpoint of making the height of center of mass of a robot low, which reduces an impact to be applied to the robot and the possibility of making the robot broken. However, the robot crawling direction has the possibility of changing due to the collapse of the surface of rough terrain. The crawling direction control aims to correcting the robot crawling direction so that it can make an error in the direction to target position small. In a simulation, we confirmed that the crawling direction control contributed to changing traveling direction on flat ground.

This paper addresses Deep Neural Network (DNN) for Sound Source Identification (SSI) of acoustic signals recorded with a microphone array embedded in an Unmanned Aerial Vehicle (UAV), aiming at people’s voice detection quickly and widely in a disastrous situation. It is well known that training a SSI-DNN needs huge dataset to improve its performance, but building such a dataset is not often realistic owing to the cost of annotation done by human. Therefore, we propose Partially Shared Deep Neural Network (PS-DNN) training using noise-suppressed acoustic signals, which can be obtained in automatic process, in addition to label data annotated by human. This results in more accurate SSI in the situation of lack of dataset for training.

Biological snakes have a soft elongated body and adapt to various environment in spite of simple form. By using them for engineering, snake robots can move in complex environment. In the case that a snake-like robot goes into pipes, it can use helical rolling motion. In previous study, a method to make a bending helical shape for a snake-like robot is proposed. However the method is not utilized in a bent pipe connected with straight pipes. In this paper, we realize going through the pipes by using helical rolling motion and bending helical rolling motion. In addition, by considering a direction of bending helical rolling motion, we propose a method to adjust a snake robot to a bent pipe. The proposed method is verified by experiments.

This paper addresses acoustic event detection and identification. These functions should be integrated into a UAV system when we consider an application such as human detection in a disaster situation. For this, we should not focus only on one of these functions, but on both of them. We, thus, propose simultaneous optimization of sound source detection and identification. Preliminary results showed that the proposed optimization can improve the total system performance compared to the case when one of theses functions is optimized.

This study is concerned with modeling and model-based controller design for hydraulic manipulator, aiming at dynamic and accurate position/force control. In general, hydraulically operated system is too complicated to be modeled by first principles. Based on a linear transfer function model obtained via system identification technique, we designed PID feedback control system and estimated efficiency of the model. We also examined accuracy of trajectory tracking control with respect to the end-effector position, under joint-wise PID feedback control.

In insect flapping mechanisms, how to hover insects' body with 4-wings flapping is still unclear part. We already demonstrated that two rotational moments could be controlled by changing the angles of the flapping amplitudes. However, the elucidations of three rotational moments for postural control with 4-wings flapping were not enough. To clarify the relationships between the aerodynamic parameters, the flapping parameters and these three rotational moments to change the moving direction while flying, we developed the test device mimicked the dragonfly. In this paper, we demonstrated our device and the results of the three moments generated by flapping motion. As the results, the roll moments could be controlled by changing the flapping amplitude angle and the pitch and yaw moments were by changing the angles of the flapping amplitude and of the stroke planes.

We have been investigating how to use protists as living micromachines by controlling their behaviors using their taxes. Because of the difficulty of attaching tools on protist such as Paramecium, we are now investigating the applicability of the water flea for bio-micromachine. Water fleas are small crustacean and they wear outer shells as external skeletons. Therefore, we can attach operational tools on the shell by chemical adhesion. In our former studies, we succeeded to control Daphnia magna's motion by illuminating the blue LED light to guide the swimming direction using their strong positive phototaxis. We tried to enhance the object pushing ability of Daphnia by the attaching twin bending needles type operational tool. However, the controllability while transporting the object was very bad and hence the total transportation time was not improved comparing to transport by Daphnia only. In this study, we tried estimate the effect of the attaching position and the size of the polypropylene rod type operational tool on the Daphnia's back to the Daphnia's swimming performance. The results showed that it was important to match the rod angle to the Daphnia's body axis (Roll and Yaw). To set the distance of the Daphnia's body center and the center of gravity of the attaching tool as short as possible is very important to improve the Daphnia's maneuverability. We developed a new operation tool called ‘Y-wing' whose structure was that a v shaped vending polypropylene rod was attached on a polypropylene straight rod. By using this new operation tool, LED movable type new motion control pool and new control program, the object transportation time along star-shaped target route was drastically reduced to one-fifth comparing to the transportation by Daphnia only. It means that the operation tool enhance the Daphnia's object transportation ability.

Three oscillating fin propulsors connected in parallel and staggered was investigated to aim an environment friendly and energy saving propulsor. In this study, propulsive speed and thrust forces were investigated. In addition, two dimensional flow structures behind the propulsors obtained by PIV technique were evaluated. As the result, staggered fin setting (front: 1, rear: 2) generate high propulsive performance. PIV detected flow pattern indicates a jet like flow pattern on staggered two and three fin settings. These jet-like flow patterns suggest existence of dominant flow patterns providing high thrust force with low power relative to conventional fin propulsors.

In recent years, the movement style of the robot have been active in various places: flying, walking on the land and swimming under the water. However, shallow waters and marshes are present in the nature. In other words, it is necessary to develop a robot with excellent behavioral in such shallow water. In this study, we focused on the water strider which has been living in the surface of the water by using the surface tension of water. By using the micro structured technology and some processing machine, the water mobile robot, was able to support the body by surface tension in the same way as the water strider.

We measured the jumping forces of butterflies Idea leuconoe during take-off. We fabricated a force plate that was composed of three MEMS (Micro Electro Mechanical Systems) force sensors and a flat plate. The three MEMS sensors were placed under the flat plate. The size of the MEMS force sensors was 7 mm×7 mm. The size of the plate was 45 mm in diameter. The resolution of the force plate was 0.20 mN. The results suggested that the maximum jumping force was 1.4 times larger on average than the gravity force. Furthermore, we revealed that the impulse of the jumping force was twice larger than the impulse of the wing aerodynamic force.

We are developing the biomimetic underwater robots using flexible piezoelectric fiber composites. The robots are inspired from structure and swimming shape of fish and can achieve high propulsion performance. To enhance the swimming performance, we introduce elastic caudal fin structure with top and bottom bones in this work. The design of elastic caudal fin and experimental results are reported.

This paper proposes a butterfly-inspired flapping-wing robot that is equipped with a variable flapping angle mechanism and describes generated force when the flapping angle was changed. Motion analysis of the mechanism showed that the flapping angle changed from 104 degrees to 129 degrees. Generated forces in vertical and horizontal directions were measured using a three-axis load cell and effect of the flapping angle on the fluid forces was investigated. The results showed that the force in vertical direction was large at a minimum flapping angle and a maximum flapping frequency.

Historically, many mobile robots for terrestrial use have either been wheeled, tracked or legged. However, some robots can locomote using body motions; not wheels, tracks or legs. These robots have been developed based on locomotor mechanism of limbless animals such as snakes. Prior work of snake robots reveals several serpentine locomotor efforts. On the other hand, there is a flatworm having flattened form and soft body such as a planarian, a polyclad and a trichoplax. The motivation for this work is the design and control of limbless robots inspired by flatworms. In this report, we propose configuration of the mechanical model of flat-formed robot. Then, we report the result of the dynamics simulation on biological or non-biological locomotion and the design of joint mechanism.

This study aims to understand relationships between musculoskeletal structure of frogs and swimming performance without control of the nerves systems. Frogs generate great swimming performance by extension of the hind limbs. Extension of knee cannot only generate thrust but also give a load to the Plantaris longus (PL) muscle, which is ankle extensor muscle. We developed an aquatic frog robot, Xenopus-noid, which can extend the knee and is driven by the dissected PL muscle from Xenopus laevis. Xenopus-noid has the knee extension mechanism composed of electrical magnet, a torsion spring positioned in hip joint. An energy of a torsion spring is released by turning off of an electrical magnet. In this study, we confirmed that the knee extension of Xenopus-noid is similar to that of X.laevis. Xenopus-noid is expected to be a useful tool to understand the relationship between knee extension of frogs and swimming performance.

We have been investigating how to use microorganisms for bio-micromachines. We succeeded to control the motion of the protists by using their galvanotaxis and phototaxis. In this study, we investigated how to control Bacillus subtilis using their chemotaxis. The difficult point to make a motion control system using chemotaxis is that once we sprayed the chemical substances for control, how to erase this effect in the next control. We thought the neutralization is suitable if we use acid/alkaline as a spraying substances. In our previous researches, we clarified the basic reaction of Bacillus subtilis to the acid and the alkaline solution. The results showed Bacillus gathered to the acid, not neutral as generally believed. We made a chemical substance spraying system by modifying the inkjet printer and succeeded to make a various character shaped bacteria groups by using this system. In this study, we investigated the movement method of bacteria group by sequentially spraying alkaline and acid. By using this method, we succeeded to move the bacteria group to our desired direction and succeeded to transport some particles(Silicon Carbide) by the bacteria group.

We proposed a vacuum gripper array for improving shaped followability. The vacuum gripper was imitated an octopus suction cup. In previous work, we developed the half spherical gripper with membrane suction cup array. The half spherical gripper can grasp the object with flat or curvature surface. However, it is difficult that the half spherical gripper grasped the object with high step. The suction cups around edge were deformed under pressing, because the suction cups are connected each other by thin elastomer. Then, we proposed the vacuum gripper has separated suction cups array. The fabricated gripper array generated highest adhesion force of 16 N when the pressing force of single suction cup was form 1.4 to 2.1 N. This gripper array can grasp the object with 10 mm in step height, in contrast to half spherical gripper which can grasp the object with 3 mm in step height.

In this paper, we investigated to drive a seesaw by automatically position controlled downward flow of the Euglena's bioconvection. The position of the downward flow was controlled by applying Euglena's phototaxis to the scanning blue laser beam. Formally, we succeeded to drive a seesaw by electrically position controlled downward flow of the Tetrahymena's bioconvection. In this paper, we aimed to reproduce the laser control version of this seesaw driving by using Euglena. We have already confirmed that if the experiment is done in the horizontal planar pool, we could make any shapes of Euglena group and we could also assemble simple mechanical parts by pushing the part by Euglena group. And in the vertical pool, controlling position of the bioconvection downward flow was done by irradiating from top surface by scanning blue laser. We succeeded to drive the seesaw by this method. In this study, we investigated the automatic laser controlling algorithm for driving seesaw. At first, we developed algorithm considering the position of the blue laser and the angle of the seesaw. Although driving the seesaw itself was succeeded, driving frequency was less than manual controlling. Therefore, we changed the algorithm that assumes the position of the bioconvective downward flow from the RGB values of the experimental images in the pool for controlling the laser position. By this new algorithm, we succeeded to drive the seesaw faster than the former algorithm and close to the driving time in manual controlling.

We conducted the motion analysis of polychaete worm, and developed the omni-directional aquatic multi-link propulsion mechanism. Movement of fins of the omni-directional aquatic multi-link mechanism was based on the motion of parapodia of polychaete worm. In this study, the width of fin corresponding to the length of parapodia were changed. We used soft elastic fins to deform easily when adjacent fins were interfered for greater width of fins. We discussed the propulsion speed characteristics with bending curve of the body and fin width.

We proposed the Octopus-bioimitating vacuum gripper with bellows structure for preventing gripper from releasing object when the object tilts. The bellows structure allows the gripper to deforming in grasping object. We evaluated flexibility of gripper on the distance between center of gravity and adhesion position. The gripper with bellows structure can grasp the 1.8 times of distance using the gripper without bellows structure. The assembly control can be simple using the robotic system with proposed gripper.

In this study, we measure the effect of the flapping by the male silkworm moth and build the periodic sniffing system for the chemical detecting robot based on the result. Recently, it has been attracting attention that autonomous mobile robots could detect an odor source by the odor information in the air. However, it is generally difficult to solve the detecting odor problem with only engineering techniques due to complexity of distribution of the odor. On the other hand, the male silkworm moth has the high ability to detect the odor source, the female correctly for breeding. It flaps its wings and sucks the sexual pheromone from the female during detecting. Building the periodic sniffing system based on moth flapping for the robot, though we succeed in improving the detecting performance compared with the case of no sniffing, there is no significant difference between with constant sniffing and with periodic.

This article deals with thrust improvement of a flapping micro-mechanism driven by external magnetic fields. The flapping wing is attached to a magnet supported by a torsion wire and produces thrust by drag difference during up and down strokes. At first, the wing with a length of 6mm was examined by varying the structure such as the wing shape and the initial horizontal angle. As a result, the thrust was greatly improved by making the wing close to an ideal motion. Next, based on this finding, the flapping mechanism with wings reduced to 4mm was fabricated and could successfully fly in low magnetic fields.

To elucidate the mechanism of bundle formation in bacterial flagella, we numerically analyzed two bacterial helical flagella rotating in close proximity. Fluid analysis reveals that the rotation generates little attraction between them. We traced the deformation process of the helices due to this hydrodynamic force, taking into account the compliance of the hook of the flagellum. At first, two model flagella, which only deform at the proximal ends, tilt according to the flow generated by the other flagellum. Then the flagella tilt like twisting each other. These flagellar behaviors are also achieved only by the anti-rotation of the cell body due to the countertorque of the flagellar rotation. The degree of the tilt and twist is comparable when the rotation rate of the cell body is 1/10 of the flagellar rotation rate, which is reasonable for actual bacteria.

In this study, we have proposed a novel magnetic control method of a swimming mechanism applicable to capsule endoscope for stomach inspection. This mechanism is composed of a magnet and an elastic plate, and is attached to a central portion of the capsule. When AC magnetic field is applied in the direction perpendicular to the water surface, the magnet is angularly oscillated by magnetic torque. Then the elastic plate swings and generates thrust in the horizontal direction. At the same time, applying a DC magnetic field in a horizontal surface, the propulsion direction of the capsule can be controlled. Furthermore, when AC magnetic field biased with DC magnetic field is applied in the direction perpendicular to the water surface, the capsule can move in the vertical direction.

We propose self-repairing mechanism using wire carried catalysts for tendon-driven mechanism. Tendon-driven mechanism can reduce weight of the arm and hand, and make robot flexibility. This mechanism is suited for the robot coexisting with human beings. But, in tendon-driven mechanism, the wire used for a long period is worn away by friction with the pulleys, eyelets and so on. To keep the tendon-driven mechanism operating over a long period of term, the self-repairing function is required for the wire of them. So, we propose the wire repairing mechanism using wire carried catalysts. The result of coating test, strength test and repairing test shows the self-repairing mechanism providing repairing agent and supporting catalysts on wire can put the long-term repairing into practice.

In this paper, a new pneumatic soft sheet actuator driven by traveling plane wave called “Wavy-sheet” is proposed. The proposed actuator consists of three rubber tube chambers and a sleeve to restrain expansion. It has three properties: 1) Generate a propulsive force by traveling plane waves 2) Flexible body 3) Sheet-shaped structure. When pressure is applied to the chambers sequentially, the actuator produces traveling plane waves and generate propulsive forces. Taking advantages of these properties, this actuator can be used for a mobile robot or for a conveyor. The relationship between the input pressure and velocity, amplitude, loads were measured to clarify the basic characteristics of the actuator.

This paper presents the improvement of structure and wing shape of the articulated wing robot that has been developed in our previous paper. Most of ornithopters currently developed deal with only flapping motion which is one of the basic wing motions of birds. Bird can fly by combining four wing motions, that is, flapping, lead-lagging, flexion and feathering, during normal flight. Flight of ornithopters which can achieve four wing motions is not realized yet. In our previous paper, we have developed an articulated wing robot which can achieve four wing motions. In this paper, we improve structure and wing shape of the articulated wing robot. By comparing thrust and lift forces of the improved robot with previous model, we show effectiveness of the difference of wing shape.

This paper described the vacuum gripper combined with electrostatic adhesion. We evaluated the electrostatic chuck using the fabrication method of MEMS. When the applied voltage was 2.4 kV, the fabricated electrostatic chuck can grasp a paper (80 mg). The adhesion force per unit area was 7.8 Pa. We first tried to fabricate the vacuum gripper with flexible electrostatic chuck. This gripper can grasp the paper (6 mg, 100 mm²) when the applied voltage was 1.6 kV.

In order to apply the ultrathin polymer film (hereafter referred to as nanosheet) as skin adhesive devices, it is indispensable to implement the wireless communication function. Accordingly, this paper shows a design method for mounting RFID on the nanosheets. In logical design, it is difficult to regulate important factors of RFID: quality factor and resonant frequency, because the resistance of the printed wire on the nanosheets is high, just over 1 [Ω/sq]. Thus, deriving the design of the highest operating possibility experimentally, all of 12 sample RFID tags work. In addition to the 12 samples, preparing several ones whose resistance is increased intentionally and whose design is not the derived one, whether tags operated or not is substantially matched to the derived operating condition.