The science of soft robots, or soft robotics, is currently one of the most active fields in robotics. While traditional robots consist of rigid bodies, powerful servomotors, and carefully coded programs to realize power, precision, and reliability, soft robots consist of soft and flexible bodies, actuators, and intelligence for adaptability. They are not rigid, but instead flexible toward their surroundings. These differences have the potential to make soft robotics a great new field in robotics.

A JSPS KAKENHI project “Science of Soft Robots” has been in progress in Japan since 2018. Part of this special issue is made in collaboration with this project. This special issue consists of 46 works in total: 2 review papers, 29 letters, and 15 papers. One review paper, 29 letters, and 3 research papers report research activities from the JSPS KAKENHI project, and the other review paper and 12 research papers have been contributed from outside the project. As this issue will make clear, the science of soft robots is a very interdisciplinary academic field, a collaboration of many researchers from various fields, such as mechanical/electrical engineering, computer science, material sciences, biology, zoology, medicine, and nursing, among others. We believe interdisciplinary work to be a key point for the exploration of soft robotics.

The editors thank all of the authors and reviewers of the contributions, and are confident that this special issue will greatly contribute to further progress in robotics.

Since 2018, a project of MEXT Grant-in-Aid for Scientific Research on Innovative Areas, titled “Science of Soft Robots: Interdisciplinary integration of mechatronics, material science, and bio-computing” has been in progress. This major research project on soft robotics in Japan has a research period of 5 years. An outline of the project is presented herein.

In recent years, conventional model-based motion control has become more challenging owing to the continuously increasing complexity of areas in which robots must operate and navigate. A promising approach for solving this issue is by employing interaction-based robotics, which includes behavior-based robotics, morphological computations, and soft robotics that generate control and computation functions based on interactions between the robot body and environment. These control strategies, which incorporate the diverse dynamics of the environment to generate control and computation functions, may alleviate the limitations imposed by the finite physical and computational resources of conventional robots. However, current interaction-based robots can only perform a limited number of actions compared with conventional robots. To increase the diversity of behaviors generated from body–environment interactions, a robotic body design methodology that can generate appropriate behaviors depending on the various situations and environmental stimuli that arise from them is necessitated. Electronics-free robotics is reviewed herein as a paradigm for designing robots with control and computing functions in each part of the body. In electronics-free robotics, instead of using electrical sensors or computers, a control system is constructed based on only mechanical or chemical reactions. Robotic bodies fabricated using this approach do not require bulky electrical wiring or peripheral circuits and can perform control and computational functions by obtaining energy from a central source. Therefore, by distributing these electronics-free controllers throughout the body, we hope to design autonomous and highly decentralized robotic bodies than can generate various behaviors in response to environmental stimuli. This new paradigm of designing and controlling robot bodies can enable realization of completely electronics-free robots as well as expand the range of conventional electronics-based robot designs.

In this letter, ostrich-inspired soft robotics, an approach to intelligent robots that can achieve dexterous manipulation and locomotion without hesitating to collide with the surrounding environment, is proposed. The rationale behind the approach is described from the history of bio-inspired mechanisms, biology, and the theory of robot control. This letter focuses on the manipulator. The first prototype of an ostrich-inspired manipulator was developed to investigate its feasibility. This prototype is a serial chain of 18 rigid links connected with rotation joints moving in a vertical plane and driven through two asymmetric antagonistic wire systems connected to two levers that are directly operated by a human operator playing the role of the controller. Therefore, this manipulator is a highly underactuated mechanism that is flexible against external forces. The experimental results show that a human operator can control this manipulator so that its tip (i.e., the head) can reach several positions, including an upper position against gravity, indicating the potential of ostrich-inspired manipulators.

Soft robotics can dramatically increase the affinity between machines and biological systems. Designing the machine/device to be soft and deformable allows the biological system to interact with the robotic system(s) mechanically, electronically, and chemically. This advantage is evident from the rapid growth of collaborative robotics, where a robot can be mechanically guided by an operator to learn motions from them without the need for coding. This letter introduces a method for combining a soft robotic system with a biological system, demonstrated through a series of case studies of ongoing research projects. These various projects have a common purpose in creating self-modifying bio-soft robots.

Flight and swimming in nature can inspire the design of highly adaptive robots capable of working in complex environments. In this letter, we reviewed our work on robotic propulsion in the air and water, with a specific focus on the crucial functions of elastic components involved in the driving mechanism and flapping wings. Elasticity in the driving mechanism inspired by birds and insects can enhance both the aerodynamic efficiency of flapping wings and robustness against disturbances with appropriate design. A flapping wing surface with a stiffness distribution inspired by hummingbirds was fabricated by combining tapered spars and ribs with a thin film. The biomimetic flexible wing could generate more lift than the nontapered wing with a similar amount of power consumption. Underwater flapping-wing propulsion inspired by penguins was investigated by combining the 3-degree-of-freedom (DoF) flapping mechanism and hydrodynamic calculation, which indicates that wing bending increases the propulsion efficiency. This work demonstrates the importance of passive deformation of both wing surfaces and driving mechanisms for improving the fluid dynamic efficiency and robustness in flight and swimming, as well as providing biological insight from an engineering perspective.

The emergence of soft robots with “flexible motion” is expected to be improved by incorporating flexible energy harvesting technology and electronic devices with excellent biocompatibility. Therefore, it is important to improve the design and performance of the device itself, according to the soft adherend to which the device is applied. In this study, we outline the design of flexible devices from a mechanical viewpoint and introduce our recent achievements.

This study aims to develop three-dimensional soft robots for special situations that cannot be easily resolved by normal hard-metal robots. We use an ion polymer–metal composite actuator, which is a soft actuator, moved using 1–10 V. The objective of our study is twofold: first, to develop a method to create 3D soft robots; and second, to develop a novel material to increase the performance of soft robots. The actuator fabricated using the proposed material exhibits a performance 113% higher than that of conventional actuators.

In this project, we aim to establish a design theory as well as implementation methods for deformable robot mechanisms that can branch and change in shape, structure, and stiffness. As the first step in our research on this project, we present an initial prototype of a branched torus mechanism that uses an inflatable structure inspired by a nemertea proboscis. We develop a basic mechanical model of this proboscis structure, and we confirm the basic performance and effective functionality of the configuration experimentally using a real prototype, specifically, a deployable torus mechanism and a retractable torus mechanism with an incompressible fluid. In addition, as an expanded concept from the branched torus mechanism, robotic blood vessels that can have an active self-healing function are prototyped, and the basic performance of the actual prototype is confirmed through experiments.

A McKibben-type pneumatic artificial muscle (PAM) is a soft actuator that is widely used in soft robotics, and it generally exhibits complex material dynamics with nonlinearity and hysteresis. In this letter, we propose an extremely durable PAM containing carbon black aggregates and show that its dynamics can be used as a computational resource based on the framework of physical reservoir computing (PRC). By monitoring the information processing capacity of our PAM, we verified that its computational performance will not degrade even if it is randomly actuated more than one million times, which indicates extreme durability. Furthermore, we demonstrate that the sensing function can be outsourced to the soft material dynamics itself without external sensors based on the framework of PRC. Our study paves the way toward reliable information processing powered by soft material dynamics.

We here introduce three types of self-actuating and nonelectronic machines using chemical reactions and physicochemical transformations. Our strategy is to develop completely artificial and autonomous machines that do not rely on electronic components. We herein demonstrate Belousov-Zhabotinsky gel machines, active droplet machines, and paper machines.

One of the goals of soft robotics is to implement intelligent functions capable of processing complex information in soft materials. This is a noble goal, and we already have a familiar example, albeit not an artificial one, in a living organism. We believe that the intelligent biological elements acquired through the evolutionary process, which do not require an electricity supply or CPU, can be used for soft robotics. In this letter, we introduce three biological elements: proteins, squid, and nematodes, which show temporal or special patterns. We then discuss an attempt to apply them to soft robotics.

In this study, we developed environmental gas sensors based on bilayer-type organic semiconductors. The number of stacked molecular bilayers was controlled through a solution-based approach. In particular, single molecular bilayers (SMBs) were produced through a geometrical frustration method that can effectively suppress the multiple stacking of bilayers. The layer number-controlled films were utilized to form thin-film transistors (TFTs) to detect the moisture in the air. We revealed that the sensitivity was enhanced in the SMB-based TFTs compared with the TFTs with thicker active layers. These findings are expected to facilitate a new route for producing flexible and lightweight chemical sensors.

In this letter, we introduce biohybrid robots powered by skeletal muscle tissue. Culturing myoblast-laden extracellular matrix structures enables the construction of skeletal muscle tissue in vitro. Biohybrid robots constructed by the integration of such fabricated muscle tissue with robot skeletons have achieved various movements, according to the configuration of the skeleton. We believe that biohybrid robots will increasingly become available in the field of robotics.

Information processing in axons has attracted attention for potential application in the control of soft robots. In this letter, we present the evaluation of axon conduction properties in cultured sensory neurons. Distal axons showed latency oscillations in response to high-frequency stimulation, thereby suggesting the suitability of our method for evaluating the information processing function of axons. Understanding axon information processing has a potential to contribute to the development of an advanced control method for soft robots.

This letter introduces a research project as invited research of Science of Soft Robots from 2019–2020. The primary purpose of this research is to investigate the mechanical durability of fabrics subjected to repeated bending, as when used as a structural part of a soft robot. The fabric is then applied to a soft robot, a gripper for food handling. Two experimental durability testing devices are developed, and the results of initial measurements are presented. Additionally, a soft pneumatic hand with membranes, inspired by the form of a deep-sea octopus, is also presented as an application example.

An important research direction in soft robotics could be the realization of environmentally friendly “green” soft robots that are biodegradable, sustainable, and recyclable. We present recent findings from our ongoing research on biodegradable robotic devices made of gelatin-based materials and discuss future directions in this contribution.

Actuators are usually driven in a uniaxial direction, which limits their ability to be driven with multiple degrees of freedom. In this study, we propose an actuator that is not limited to a uniaxial direction. We developed a polyvinyl chloride gel actuator with a particulate structure. The actuator can change its surface shape by displacing each particle in the structure. As the first step in this experiment, each particle was displaced independently, by applying a voltage to the anode to change the actuator’s surface into an uneven shape.

This letter proposes a peristaltic mixing pump based on bowel peristaltic motion driven by soft pneumatic actuators. Furthermore, various practical applications using the peristaltic mixing pump are introduced, such as mixing and conveying processes for solid rocket fuel production and lifting conveyance of excavated soil and sand during the construction process. First, the bowel peristaltic motion mechanism is introduced from the anatomical perspective. Next, the construction of the peristaltic mixing pump and its actuation mechanism are presented. Finally, the application and prospects of peristaltic mixing pumps are described.

The implementation of autonomous functions, such as autonomous actuation, self-healing, and learning functions, has been a potent strategy to realize adaptation abilities against changes in environments and sudden incidents. Organic materials, such as living cells and tissues, can be used as robot parts for the implementation of autonomous functions because they can modify biological functions and remodel tissue morphologies in response to the environment. A brain organoid is a cell aggregate formed by recapitulating the development processes of the fetal brain in vitro. Because the brain organoid reproduces complex 3D structures and various cells, it can be used as a living regulator of robots for implementing complex autonomous functions. In contrast, engineered muscle tissues constructed by culturing myoblasts with biomaterials can also be used as a living actuator for robots. Therefore, to implement autonomous functions for robots, we have proposed methods for connecting the brain organoid with engineered muscle tissue and for co-culturing complex in a culture vessel.

This study focuses on the mechanical contact between a soft body and the environment, referred to as soft contact. First, the soft contact between an object and a soft capacitive sensor was analyzed. We derived the stress-strain relationship of the dielectric in a soft capacitive sensor, where the sensor output depended on the force; however, it was independent of the contact area. Second, the soft contact between an elastic jumping robot and the environment was analyzed. We derived the local optimal shapes for higher jumping.

A number of recent studies have exploited the sizes and functional properties of microdevices and cellular mechanical components to construct bio-microdevices. As the scale of microdevices can accommodate different cell sizes and processing capabilities, a number of efficient bioreactors and bioassay systems using cellular functions have been produced. To date, the main focus of these devices has been the analysis of cellular chemical functions. On the other hand, our concept is to use cells as components of devices for fluidic control. To date, various devices have been developed that exploit cellular mechanical functions. The working principle of these devices is novel because they only use chemical energy inputs. In this letter, the recent progress of this study and its characteristics are reviewed.

This letter proposes a novel therapeutic approach in assisted reproductive technology (ART) to control the implantation position of after embryo transfer. The system composed of a soft microrobot, a catheter, and a guiding magnet. The microrobot accommodates and transports an embryo into the patient’s uterus and keeps the embryo within the suitable area for implantation. The proposed treatment was demonstrated with the prototype in an obstetric model. This minimally invasive system will increase the pregnancy rate and prevent ectopic pregnancy.

This letter describes the social needs and research trends of soft robots for pressure ulcer prevention, and our current work on the robotization of soft gel cushions.

In this letter, we review the results of our recent studies on a soft microswimmer powered by fluid oscillations. The microswimmer consists of an elastic membrane with a prolate spheroidal reference shape containing a rigid sphere. The swimming direction can be controlled by appropriately applying fluid oscillations. The obtained knowledge will be useful for future artificial microswimmer designs.

Insects exhibit resilient and flexible capabilities allowing them to adapt their walk in response to changes of the environment or body properties, for example the loss of a leg. While the motor control paradigm governing inter-leg coordination has been extensively studied in the past for such adaptive walking, the neural mechanism remains unknown. To overcome this situation, the project “Motion Hacking” develops a method for hacking leg movements by electrostimulating leg muscles while retaining the natural sensorimotor functions of the insect. This research aims to elucidate the flexible inter-leg coordination mechanism underlying insect walking by observing the adapting process of inter-leg coordination with the insect nervous system when leg movements are externally controlled via motion hacking.

Cursorial quadrupeds have different connections to the trunk for forelimbs and hindlimbs: a flexible connection through the muscles to the forelimb, and a secure connection through the hip joint to the hindlimb. Although anatomical and biological studies have described the structure and behavior of cursorial quadrupeds by focusing on flexible shoulders, the functionality of the flexible shoulder remains unclear. In this study, we first survey the anatomical and biological studies. Second, we introduce our robotics studies, which focus on flexible connections for proximal limb joints. Further, we discuss future directions for extracting a design principle based on complex animal body structures, and we suggest the potential for interdisciplinary research between anatomy and soft robotics.

Structural changes during the drawing and shrinking of linear low-density polyethylene (LLDPE) film are analyzed through in-situ X-ray measurements. A synchrotron radiation source enables simultaneous analyses combining small-angle X-ray scattering (SAXS) and wide-angle X-ray diffraction (WAXD). During drawing, the original unoriented SAXS and WAXD patterns transformed into line and spot patterns, indicating the orientation of both lamellar stacking and the molecular axis along the drawing direction. During the subsequent shrinking these patterns are retained, suggesting the tilted lamellar and molecular chains. A possible model for structural changes indicates that tie molecules between lamellae effectively transmit drawing and shrinking stresses, which contributes to desirable actuation properties.

A neural interface is a technology that facilitates communication between the brain and external devices. One potential clinical application of a neural interface is in prosthetic arms. These devices can realize motor and sensory functions, enabling amputee patients to perform daily tasks. However, such prosthetic arms are still challenging because of the poor resolution of conventional neural electrodes and the difficulty in sorting the motor and sensory nerve cells. In this study, we attempt to utilize deoxyribonucleic acid (DNA) nanotubes to develop a novel intracellular electrode with high resolution. Our results indicate that DNA nanotubes can transport ions, such as Ca²⁺. Moreover, a microchannel device for sorting motor and sensory nerves is introduced.

Long before reaching its mechanical breaking point, a bio-system begins responding to stress at its own “bio-functional breaking point,” a phase of life activity dysfunction. However, little is known about the correlation between tissue flexibility and the conditions under which cellular response, damage, and death occur. We are now developing a new confocal microscopy-based observation method to analyze cell aggregates (spheroids) that are under physical pressure. The method concomitantly assesses cellular responses, stress levels, and cellular structure changes. Using this method, we found that the artificial suppression of the gene expression of fibronectin, a major component of the extracellular matrix, provides different mechanical characteristics to hepatoma-derived cell line spheroids than does the control wild type. This study may aid in the prediction of the characteristics of a tissue of interest by simply analyzing the tissue gene expression pattern, providing valuable information for the development and operation of wearable devices. It may also help in the preparation of custom devices that suit specific individuals.

For in-vehicle optical communication systems, an air gap exists between two plastic optical fibers (POFs) to avoid damage to the POF end due to vibrations. There exists the disadvantage that a loss occurs in air gaps. To solve the issue of high-loss optical coupling, in this study, we designed a light-induced self-written (LISW) optical waveguide between POFs using gel material. It was demonstrated that the two optical fibers were automatically interconnected through the LISW optical waveguide, and the connection maintained the adhesiveness against mechanical displacement.

With the increase in online interaction between distant people, wearable avatar robots are expected to be used in a wide range of daily situations. Soft robotics is highly applicable as a method to achieve this. In this study, we define the design requirements for the daily use of wearable soft avatar robots based on design surveys and cross-field academic research. In addition, we implement prototypes using an inflatable robot and summarize the future issues.

We propose a wire-driven robotic arm inspired by the ostrich neck. It can pick up a small piece of feed from the ground while colliding with it. This arm is named robostrich arm (shortened form of robotic ostrich arm). It consists of a serial chain of 18 rigid bodies connected by free rotational joints that are designed to have angle limitations similar to the bones of a real ostrich. It moves in a vertical plane and is driven by two DC motors through antagonistic wires. The task considered in this study was to lift the arm tip (the “head” of the robostrich arm). The experimental results indicate that the tensioner balance and timing between the two wires are important for achieving the head-up task. This paper indicates the contribution of antagonist muscles to the performance of head-up tasks by high-degree-of-freedom underactuated manipulators in robotics and ostrich necks in biological studies.

The coronavirus epidemic has attracted significant attention to the applications of pet robots which can be used to treat and entertain people in their homes. However, pet robots are fabricated using hard materials and it is difficult for them to communicate with people through contact. Soft robots are expected to realize communication through contact similar to that of actual pets. Soft robots provide people with a sense of healing and security owing to their softness and can extract rich information through external stimuli by applying a machine learning framework called physical-reservoir computing. It is crucial to determine the differences between the physical properties of soft materials that affect the information extracted from a soft body to develop an intelligent soft robot. In this study, two owl-shaped soft robots with different softnesses were developed to analyze the characteristics of the signal data obtained via piezoelectric film sensors embedded in models with different physical properties. An accuracy of 94.2% and 95.9% was obtained for touched part classification using 1D CNN and logistic regression models, respectively. Additionally, the relationship between the softness of material and classification performance was investigated by comparing the distribution of part classification accuracy for different hyper-parameters of two owl models.

In recent years, untethered soft robots, free of the lines that restrict their mobility, have been studied extensively. Our research team has been focusing on the electrohydrodynamic phenomena (EHD) as a driving mechanism for untethered robots. EHD is a phenomenon in which a flow is generated by applying a high voltage to a dielectric liquid. We propose a method to drive a robot in an untethered manner using EHD by vertically stacking two types of liquids: conductive and dielectric. This method is simpler, more energy-efficient, and quieter than conventional systems. Although a lower voltage would prevent the enlargement of the system by limiting the electronic components, the generation of EHD requires a high voltage. Therefore, in this study, to realize the low voltage drive of untethered robots dominated by the electrostatic actuator, we tackled the reduction of the driving voltage by investigating the phenomenon. As a result, we achieved low voltage driving at 15 V and successfully drove with off-the-shelf batteries (18 V). We also investigated the output current flowing through the system to reduce power consumption. Therefore, in addition to improving the energy efficiency of the system, we confirmed that the difference of the generated current depended on the thickness of the dielectric liquid and the concentration of the conductive liquid.

Conventional mobile robots typically use a wheel or crawler mechanism for locomotion. However, these robots often get stuck in narrow spaces. To mitigate this issue, we propose an innovative flexible mobile robot named “MOLOOP” in this study. The proposed robot has double-looped and hermetically-sealed structure and all ground-contact-area moves to the same direction. Three synchronized flexible crawlers generate the driving force within the robot body. The crawler, named “HS crawler,” comprises of loop-connected flexible bags. The entire body of the robot is flexible; therefore, the robot can pass through narrow terrains with adaptive shape changes. However, in terms of power and size, performance improvement is necessary for the practical use of the robot. This paper describes the improvement of HS crawler performance, which directly affects the driving force, traveling speed, and size of the robot. For the purpose of high input pressure, this study reinforced the flexible bags of the crawler using fibrous materials. This reinforcement realizes the power-up and downsizing of the flexible bags. The new crawler achieves a performance of 3.43 Nm torque and 39.0 mm/s traveling speed, compared to a torque of 0.23 Nm and traveling speed of 3.0 mm/s of the previous crawler. Additionally, the width of the new MOLOOP changed to 270 mm from 420 mm because of the downsized crawler. Furthermore, a dedicated mechanical valve was developed. This valve achieves a flow rate of 2.39 L/min as compared to 0.92 L/min of a commercially available solenoid valve. Moreover, the traveling speed of MOLOOP increased from 3.0 mm/s to 9.0 mm/s. The prototype robot successfully passed through a narrow terrain with a minimum width of 200 mm (robot width: 270 mm).

Recently, several pneumatic soft actuators have been applied to wearable and welfare devices to provide nursing care and physical support for the elderly and disabled. In this study, as a wearable soft actuator for holding body, a spiral shaped soft holding actuator that can wrap a user according to their body shape was proposed and tested. The construction and operating principle of the tested soft actuator with circumferential restraint mechanism using three extension type flexible pneumatic actuators (EFPAs) has been discussed. As a result, it was found that the tested actuator could hold elbows and knees when the joint is in motion. An analytical model of the spiral actuator was also proposed to achieve an optimal design. It can be confirmed that the proposed analytical model can predict the shape of the actuator when various EFPAs are pressurized.

Voluntary rehabilitation at home helps to prevent the joint contracture after medical treatment. Our previous studies concerned a low-cost portable rehabilitation device using a flexible spherical pneumatic actuator as a passive exercise device. However, the device requires a bulky compressor to drive it. This study results in a compact fluidic driving system that is highly flexible. The system adopts a flexible electro-hydraulic cylinder driven by an electric motor and a hydraulic gear pump. An empirical equation for the suitable pump rotation for the desired displacement of the system has been determined. As a result, the multi-position control of the system within the tracking error of 4 mm has been realized by using the on/off control scheme based on the obtained equation. In addition, a flexible, spherical electro-hydraulic actuator using two proposed drive systems is developed and tested. Control of the attitude of the tested spherical actuator is successfully realized.

The Japanese agriculture industry, faced with the problem of declining and aging farmers, has keenly called for the development of body assist orthoses for aiding agricultural workers. Therefore, in this study, we propose an endoskeleton-type knee joint assist orthosis for assisting knee joints in half-sitting postures or crouching postures. The proposed endoskeleton-type knee joint assist orthosis uses two McKibben-type artificial muscles per leg and four in total, placed from the waist to the foot along the forward part of the leg to generate assisting power. With the proposed orthosis in use, the contraction forces generated by the McKibben-type artificial muscles are converted into torque via the knees to assist the knee joints. This paper first presents the appearance of the prototyped endoskeleton-type knee joint assist orthosis, and then describes its components and the McKibben-type artificial muscles. To verify its actual effects on muscles, we conducted fitting experiments using a surface myoelectric electrometer to determine its assisting effects; these effects were found on certain experimental subjects, but there were no tangible effects on others. The experimental results suggest that the prototyped endoskeleton-type knee joint assist orthosis requires further improvement.

Fabricating a soft robot using conventional molding methods is difficult and time-consuming. Moreover, the types of materials used in the process are limited, and the elasticity cannot be changed incrementally. In this paper, we explain the detailed process of manufacturing molds for silicone joints. We construct a prototype molded silicone joint. We measure the elastic modulus of this joint and confirm that the elastic modulus and anisotropy change depending on the density, size, and arrangement of the surface grooves in the mold. We also develop a prototype human phantom using the proposed joint. We aim to contribute to the medical field by applying new techniques made possible by soft robotics.

Conventional model theories are not suitable to control soft-bodied robots as deformable materials present rapidly changing behaviors. Neuromorphic electronics are now entering the field of robotics, demonstrating that a highly integrated device can mimic the fundamental properties of a sensory synaptic system, including learning and proprioception. This research work focuses on the physical implementation of a reservoir computing-based network to actuate a soft-bodied robot. More specifically, modeling the hysteresis of a shape memory alloy (SMA) using echo state networks (ESN) in real-world situations represents a novel approach to enable soft machines with task-learning. In this work, we show that not only does our ESN model enable our SMA-based robot with locomotion, but it also discovers a successful strategy to do so. Compared to standard control modeling, established either by theoretical frameworks or from experimental data, here, we gained knowledge a posteriori, guided by the physical interactions between the trained model and the controlled actuator, interactions from which striking patterns emerged, and informed us about what type of locomotion would work best for our robot.

This paper proposes a drive principle that aims to improve the control characteristics of a particle-excitation flow control valve capable of continuous air flow control. Aiming at application to a small servo valve, the authors have developed a particle-excitation flow control valve that controls flow rate by separating particles, which act as a valve element, from an orifice that is opened by the oscillation of a piezoelectric element. This paper proposes a method of more finely adjusting the motion of the particles. This method makes it possible to adjust to what degree each orifice opens, thereby refining the control of the flow rate and improving the responsiveness of the valve. Here, the authors produce a prototype, evaluate its characteristics, and confirm its effectiveness.

We propose a vacuum end effector with an expansion and contraction mechanism to realize a picking task for objects placed in a narrow space, such as a shelf. The proposed expansion and contraction mechanism consists of a tube and exoskeleton structure and is characterized by the use of a thin metal plate wound about itself to form a tubular exoskeleton. Expansion and contraction motions were realized by connecting the tube to a linear motion mechanism. The expansion and contraction mechanism can be easily extended by elastic force. In addition, the shape of the expansion and contraction mechanism is created by winding a thin metal plate around a predetermined axis, which ensures high rigidity even in the extended state. Even when an object to be picked from a shelf is located behind other objects, the end effector can efficiently hold the object because of its elongated shape and ability to freely change the position of the suction pad using a direction-changing linkage and the expansion and contraction mechanism. The developed end effector weighs about 1.46 kg and can carry a load of 0.56 kg when extended to 150 mm. Verification of the mechanism confirmed that the developed end effector is useful because it can perform the expected object-picking operation.

Pneumatic soft actuators exhibit both passive flexibility from the casing and active flexibility from pressurizing and depressurizing. These actuators are expected to be human-friendly and are often used in nursing and medical situations because they allow access to soft systems through control of the internal air pressure. The current design of pneumatic soft actuators has two technical points of interest: the control method for the drive direction and the operating pressure level. An actuator that addresses these points is required. In this study, a pneumatic soft actuator with a silicone rubber casing – called a sponge core soft rubber actuator (SCSRA) – is developed to solve these problems. SCSRAs can perform various functions by changing the bonding state of the silicon film. Thus, a large stroke can be achieved in a low-pressure area of ≤ 30 Pa, and the driving system can be controlled by peeling off and bonding the silicon film. We clarified the expansion and stiffness characteristics of the “bonded SCSRA” and “peeled SCSRA” when unpressurized and pressurized and measured the grip strength when walking in shoes with protrusions on the insole and single-tooth sandals with protrusions on the sole as example applications of the sensing ability.

Recently, myoelectric hand prostheses produced by the combination of 3D-CAD and printer have gained attention. 3D printing of hand prosthesis has resulted in cost reduction. However, when an electric actuator with reduction gears is used as the driving source of the hand prosthesis, the joint rigidity becomes high; therefore, compliance control is required to grasp soft target objects. In this study, we propose a pneumatically driven hand prosthesis using a flexible bellows actuator. The hand prosthesis is lightweight and inexpensive because it is self-powered and generates compressed air through the supination motion of the user’s forearm instead of an external compressor, which is essential for conventional pneumatic systems. Stable flexible grasping of the target object was achieved by driving a five-finger hand using this system.

In this study, we developed a variable-stiffness and deformable link using shape-memory material and the jamming transition phenomenon. Above its glass transition temperature (T_g), a shape-memory polymer (SMP) can be deformed by applying a small load. SMPs maintain the deformed shape after they have been cooled below T_g, and they return to their original shape when heated above T_g. The reversible change in the elastic modulus between the glassy and rubbery states of SMPs can be on the order of 100–1000 times. We exploited the characteristics of SMPs to develop robot components with variable stiffness and sensitivity. The jamming transition phenomenon for granular material has been widely used as a method to change the stiffness of robots. This phenomenon is the change from fluid-like to solid-like conditions by removing air from a space containing particles. In this study, we developed a variable-stiffness link by combining the SMP and the jamming transition phenomenon. Moreover, by replacing the SMP with shape-memory alloys (SMAs), whose recovery force and elastic modulus are larger than those of SMPs, we prepared a second prototype with variable stiffness. We evaluated the performance of both prototypes, using the SMP or the SMA, with experiments and confirmed the motion principle of the proposed link (e.g., shape recovery and shape fixity). Moreover, it was confirmed that the stiffness of these links can be changed among four states.

In this study, we developed new bending modules and pneumatic self-excited valves. In a previous study, we developed a planar flexible-deformation mobile robot that combines multiple bending modules. The robot moves by a traveling wave, which is periodically generated by pressurizing air to the bending modules. Further, it can move in every direction by simultaneously combining the two directional traveling waves. However, problems such as air leakage from the bending module that constitutes the moving object and the influence of the moving operation owing to the tubes comprising the flow path were identified. To address these problems, new bending modules have been developed. To address air leak issues, we verified the materials used in the bending modules and developed an injection mechanism. To reduce the number of externally connected tubes, we developed new bending modules that could include multiple internal flow paths. In addition, because the mobile robot moves owing to the generation of traveling waves, it is a low-grade operation. However, this requires multiple solenoid valves and electronic circuits. This is useful if traveling waves can be generated by simply supplying air using a pneumatic self-excited valve. In this study, we developed bending modules with multiple internal channels and a pneumatic self-excited valve that utilizes the snap buckling of leaf springs used in the mobile robot.