Keywords: Modeling, Control, and Learning for Soft Robots, Sensor-based Control, Underactuated Robots
Abstract: This paper presents a state-space description using planar discrete elastic rod theory of a soft robotic appendage with torque input at one end. We design a linear output feedback controller to balance the appendage in an unstable vertical configuration with an angle sensor and torque input co-located at the base. Gains are tuned through simulations of the nonlinear system and hardware experiments are performed to verify performance. Simulation results suggest that the resulting control design balances some appendages that would otherwise buckle under their own weight.

Keywords: Soft Robot Materials and Design, Modeling, Control, and Learning for Soft Robots, Soft Robot Applications
Abstract: Pneumatically operated soft robots require complex infrastructure for their operation: microcontrollers must control hard pneumatic valves via power electronics. Although soft digital logic gates based on soft valves have been demonstrated as a replacement for electronic control, the development of memory from logic gates is cumbersome (three logic gates with mono-stable membranes for the development of a single S-R latch), and such memory is only capable of holding, but not storing, information; after a power reset, the membranes relax to their idle states, and the information is lost. In this work, we introduce a soft memory device with a bistable membrane that allows the permanent storage of binary information in soft materials, and we demonstrate its writing and erasing operations. We also introduce a new type of pneumatically-driven soft display, the soft bubble display. We connect the display to our soft memory device to visualize the information that is held in the memory. Our work highlights the importance of material-based memory and its future use for programming soft robots.

Keywords: Soft Robot Materials and Design, Hydraulic/Pneumatic Actuators
Abstract: In this paper we present a hydrogel with self-healing capabilities and its application for the development of a Pneumatic Artificial Muscle (PAM). Unlike other hydrogels, our material can be used outside of aqueous environments and does not need any external stimulus to self-heal, which makes it an interesting alternative for the manufacturing of soft robots. First, the mechanical properties of the hydrogel and its self-healing ability are analyzed. Second, we present the development of a pneumatic muscle based on the classic McKibben design but including our material. Finally, we analyze the capabilities of our self-repairing muscles before and after being punctured. The results show a good performance of our actuators even after low healing periods (10 minutes).

Keywords: Soft Robot Materials and Design
Abstract: Herein, we focus on the design method of a robot, named Wave Wheel, capable of generating a smooth continuous peristaltic wave, which is driven by a bundled rotary helix drive mechanism. Wave Wheel mainly consists of a braided mesh tube, multiple helices that are arranged on the circumference, and spur gears. When a single motor rotates the helices, the wheel generates peristaltic waves. The proposed mechanism has some unique characteristics: (i) In principle, smooth peristalsis can be generated with a simple structure. (ii) It can be driven by a single motor and can propagate waves at high speed by infinite rotation of the shaft. (iii) The structure is a circle in the transversal plane and can be used as an omnidirectional drive wheel. The basic design method, such as waveform, velocity, and the collision condition are discussed from a geometrical point of view. Based on the model accounting for the mechanical constraints, we have designed and fabricate a prototype robot and experimentally tested it (see movie). The prototype (diameter of 57 mm) reached the top speed of the peristaltic locomotion of 43 mm/s when angular velocity of the helix was 60 rad/s. We obtained the trajectory of the mesh surface by motion capture, and the result showed that the velocity was not constant on the whole surface but periodically changed with time due to the sliding between the mesh and the helix.

Keywords: Soft Sensors and Actuators, Modeling, Control, and Learning for Soft Robots, Hydraulic/Pneumatic Actuators
Abstract: This paper presents a novel pneumatic artificial muscle (PAM) that is able to generate a high payload and buckling pressure. The proposed actuator exploits open-cell foam, and 0.8 mm thick rigid rings to reinforce structure stiffness, and to implement stable contraction under depressurization. All components are embedded in a sealing elastomeric skin. The actuator can support 6.8 N of external load, which is 34.5 times greater than its own weight (20g). When depressurized, the actuator deforms stably, without buckling. The results of the preliminary experimental analysis show that it is able to contract up to 51.8% upon -80 kPa, with a 5.6% hysteresis for pressure vs. contraction ratio. Moreover, a quasi-static model is proposed to estimate the actuator blocking force, of which the measured maximum value is 32.7 N at -80 kPa vacuum. The presented actuator shows repeatable and reliable performance, providing a promising soft material solution for pneumatic driven movement.

Keywords: Soft Sensors and Actuators, Force and Tactile Sensing, Modeling, Control, and Learning for Soft Robots
Abstract: In this paper we present NeatSkin, a novel artificial skin sensor based on electrical impedance tomography. The key feature is a discrete network of fluidic channels which is used to infer the location of touch. Change in resistance of the conductive fluid within these channels during deformation is used to construct a tomography. We present a method to simulate touch using this unique network-based, low output dimensionality approach, the efficacy of which is demonstrated by fabricating a NeatSkin sensor. This paves the way for the development of more complex channel networks and a higher resolution soft skin sensor.

Keywords: Grippers and Other End-Effectors, Soft Robot Materials and Design, Soft Robot Applications
Abstract: Robotic food handling is a complicated task to perform when moved ingredients are delicate. It is really import to design end-effectors that can meet adaptation and reliability while limiting damage to the manipulated items. We present in this paper a novel soft-rigid gripper that exploit the combination of specialised fingertips with a passively compliant structure. The proposed device can gently grasp object of different shape by scraping below the object four spatula parts that can then lock together so to form a stable plate surrounded by the fingers. This solution allows to dramatically reduced the forces applied directly onto the object during transportation, increment the grasping success rate compared with classical gripper and reduce the precision needed to represent the object to grasp location. We have tested a prototype in a real experimental setup using real ingredients. The encouraging results suggests that this approach to ingredient manipulation may result particularly useful when delicate objects have to be moved in the context of an automated food production line.

Keywords: Neurorobotics, Cellular and Modular Robots, Biologically-Inspired Robots
Abstract: We propose a minimal yet highly biomimetic hierarchical controller based on a neuromorphic spiking central pattern generator (CPG) consisting of twelve simulated neurons modulated by sensory feedback. The robotic application of this controller is a flexible modular robot which uses DC linear actuators to morph its structure and achieve a crawling gait.

It was shown in previous work that different gaits can be achieved by altering the number and physical location of structural modules. The proposed controller enables a modular style of controller design as well. Additional behaviors, or adaption of the controller to a different flexible robotic configuration, can be achieved by additional independent "neural modules" consisting of spiking neurons.

Keywords: Grippers and Other End-Effectors, Soft Sensors and Actuators, Soft Robot Materials and Design
Abstract: While soft fingertips have shown significant development for grasping tasks, its ability to facilitate the manipulation of objects within the hand is still limited. Thanks to elasticity, soft fingertips enhance the ability to grasp soft objects. However, the in-hand manipulation of these objects has proved to be challenging, with both soft fingertips and traditional designs, as the control of coordinated fine fingertip motions and uncertainties for soft materials are intricate. This paper presents a novel technique for in-hand manipulating soft objects with precision. The approach is based on enhancing the dexterity of robot hands via soft fingertips with tactile sensing and active shape-changing; such that pressurized air cavities act as soft tactile sensors to provide closed-loop control of fingertip position and avoid object's damage, and pneumatic-tuned positive-pressure deformations act as a localized soft gripper to perform additional translations and rotations. We model the deformation of the soft fingertips to predict the in-hand manipulation of soft objects and experimentally demonstrate the resulting in-hand manipulation capabilities of a gripper of limited dexterity with an algorithm based on the proposed dual abilities. Results show that the introduced approach can ease and enhance the prehensile in-hand translation and rotation of soft objects for precision tasks across the hand workspace, without damage.

Keywords: Modeling, Control, and Learning for Soft Robots, Soft Robot Applications
Abstract: A circular elastic robot, which can move and jump by deforming its outer shell was studied in this paper. Previous research has confirmed that the initial deformed shape of the robot significantly affects its jumping height. However, the deformed shape needed for the highest jump has not been confirmed. Our goal is to optimize the initial deformed shape of the circular elastic shell. To achieve this, first, we analyze the relationship between the jump height and the initial deformed shape using a simple model. In this paper, the circular elastic shell is a discrete model approximated to eight links and joints. The deformed shape of this discrete model is optimized, and the effect of the optimized deformed shape is discussed.

Keywords: Manipulation Planning, Modeling, Control, and Learning for Soft Robots, Underactuated Robots
Abstract: Many soft robotic systems for manipulation have been developed for their compliance and their smooth obstacles avoidance capabilities contrary to rigid systems. In order to increase human task aid, a new class of robots known as mobile soft manipulators has thus emerged, where the path planning of the whole system, both soft manipulator and its mobile plate-form, remains a real scientific challenge. For that purpose, in this paper, we suggest a curvature control algorithm for that class of hyper-redundant mobile soft manipulator approaching and handling objects with simultaneous obstacle avoidance based on potential field of velocity. The algorithm is based on principle of sliding mode, which converges in a finite time. The soft manipulator kinematics is modeled in 2D by the mean of Pythagorean Hodograph (PH) quintic curve theory with length constraints. The result of this modeling and control is applied on Festo Robotino-XT, mobile and soft manipulator robot

Keywords: Force Control, Force and Tactile Sensing
Abstract: This work presents a soft robotic module that can sense and control contact forces. The module is composed of a foam spring encapsulated by a pneumatic bellow that can be inflated to increase its stiffness. Optical sensors and a light source are integrated inside the soft pneumatic module. Changes in shape of the module lead to a variation in light reflectivity, which is captured by the optical sensors. These shape measurements are combined with air pressure measurements to predict the contact force through a machine learning model. Using these predictions, a closed-loop control of the contact force was implemented. The modules can be applied to realize pressure distribution control in support devices such as seats and mattresses. The presented method is robust and low-cost, can measure both shape and contact force, and does not require (rigid) sensors to be present at the movable contact interface between the support device and the user.

Keywords: Soft Sensors and Actuators, Soft Robot Applications, Biologically-Inspired Robots
Abstract: The ability to move on an unstructured terrain and in confined spaces greatly increases the number of tasks terrestrial robots can carry out. We present a soft sensorized foot module for improving adaptable interactions with different surfaces. This design enables force measurements with repeatable and reliable information on flat, unstructured, and inclined surfaces. By characterizing the foot, we investigated the force interaction of the actuator with variable surface in three different conditions: i) normal loading on flat surface, ii) normal loading with obstacles, and iii) tangential force. Locomotion experiments were therefore conducted on a flat and an inclined surface to validate the attachment/detachment force in the normal and tangential direction. This study not only opens up a new way of achieving adaptable soft robotic locomotion but also provides a better understanding of the functionalities and sensing capabilities of the robot on anisotropic surfaces. Understanding such behaviour and capabilities may help in developing a new type of locomotion system for real-world applications.

Keywords: Soft Robot Materials and Design, Soft Sensors and Actuators, Compliant Joint/Mechanism
Abstract: One way biological creatures exploit their soft tissues and muscles is to quickly adjust to their environments in a flexible manner. Artificial soft material actuators have been inspired by these complex motions and studied to mimic or even exceed these biological systems. A widely adopted soft actuation method is by controlling pressure changes inside a hollow structure through the use of pneumatic systems, where inflating the structure would cause it to deform in different directions depending on its geometrical design. Instead, this work investigated actuation by vaporizing embedded liquid in a soft structure. In particular, ultrasonic waves can be directed to excite the structure that is partially filled with a small amount of ethanol, whereby atomization ejects small droplets of vaporized ethanol into the chamber to inflate the overall structure. As compared to boiling, evaporation by ultrasonic atomization is much faster and occurs at a lower temperature. Furthermore, tethered tubes are unnecessary, since ultrasonic waves can propagate through the structure. Here, soft actuation through ultrasonic atomization was analyzed by comparing the results from both experiments and numerical models. In addition, a soft robotic gripper was designed and fabricated to hold onto various objects.

Keywords: Soft Sensors and Actuators, Force and Tactile Sensing, Soft Robot Materials and Design
Abstract: Soft tactile skins can provide an in-depth understanding of contact location and force through a soft and deformable interface. However, widespread implementation of soft robotic sensing skins remains limited due to non-scalable fabrication techniques, lack of customization, and complex integration requirements. In this work, we demonstrate magnetic composites fabricated with two different matrix materials, a silicone elastomer and urethane foam, that can be used as a continuous tactile surface for single-point contact localization. Building upon previous work, we increased the sensing area from a 15 mm² grid to a 40 mm² continuous surface. Additionally, new preprocessing methods for the raw magnetic field data, in conjunction with the use of a neural network, enables rapid location and force estimation in free space. We report an average localization of 1 mm³ for the silicone surface and 2mm³ for the urethane foam. Our approach to soft sensing skins addresses the need for tactile soft surfaces that are simple to fabricate and integrate, customizable in shape and material, and usable in both soft and hybrid robotic systems.

Keywords: Soft Sensors and Actuators
Abstract: As a recently developed artificial muscle, supercoiled polymer (SCP) actuators exhibit many desirable properties such as inherent compliance, large linear tensile actuation, and high energy and power densities. By embedding linear SCP actuators into flexible beams and activating them, bending SCP actuators can be created with outstanding deflection performances. However, their full potential as bending robotic muscles is challenged by their hysteresis nonlinearity and the complex coupling between the beam and the linear SCP actuator. In this study, we propose a nonlinear model that can accurately capture and estimate the steady-state displacement of the bending SCP actuator. The proposed model is constructed by coupling a Preisach hysteresis model and nonlinear beam equations. Numerical solution is obtained using a bisection technique-based algorithm and Runge-Kutta method. For comparison purposes, a linear model is also considered. Both simulation and experimental investigations are conducted. The effectiveness of the proposed modeling approach is confirmed.

Keywords: Soft Sensors and Actuators, Soft Robot Materials and Design, Soft Robot Applications
Abstract: Optical sensors have recently been proposed for sensing bending, linear strains and external forces in soft actuators. However, the susceptibility of these sensors to a number of stimuli makes it difficult to comprehend the resulting data output. We address this challenge through the introduction of a low-cost flexible bending sensor based upon the abrasion of optical channels. Such sensors are responsive to bending while being insensitive to internal pressure or external forces. The developed sensor is integrated into a soft pneumatic finger in order to obtain feedback of its curvature. We further demonstrate that the sensor is insensitive to normal forces applied on a blocked finger. Through this work, we propose a simple, low-cost sensor development technique towards addressing the challenge of decoupling forces and deformations in soft robots.

Keywords: Soft Robot Applications, Soft Sensors and Actuators, Wearable Robots
Abstract: Muscle weakness following a stroke, spinal cord injuries or age can make people sedentary temporarily or permanently. The best strategy for such patients’ recovery is to motivate them to break their wheel/chair bound condition and attempt independent motion. The ability to easily transition from sit-to-stand (STS) would encourage further mobility. Hence, there is a need for wearable assist devices that seamlessly assists STS transition. Such a device should be non-obstructive during the seated phase and assist during STS. Soft exosuits are advantageous over conventional rigid exoskeletons in this context. The use of exosuits in STS assistance is currently limited by the lack of low-profile soft actuators with high strain rate and force-to-weight ratio. Hence, we propose a novel low-profile vacuum actuator that, can be rapidly fabricated, is lightweight (14 g), and can provide high strain (~65%) and a high force-to-weight ratio (~160). The proposed actuator comprises a low-profile spring encased within a low-density polyethylene film with rapid vacuum actuation and passive quick return. The applicability of the proposed actuator in assisting STS is preliminarily studied via a prototype exosuit. Surface electromyography (sEMG) measurements of the gluteus maximus (GM) muscles during STS experimentally show the potential use of the proposed actuator in an STS-assist exosuit. Experimental sEMG results indicate a mean reduction of ~46% muscle activity of GM muscle during STS.

Keywords: Soft Sensors and Actuators
Abstract: Polypyrrole (PPy) has been widely used as an electro active polymer actuator owing to its high energy density, and the applicability in low voltage (i.e. less than 3 V) driven systems. Its scalable fabrication process also enables a PPy actuator to be used in mesoscale (a few millimeter to centimeter) robotic mechanisms. Although many PPy actuator driven mechanisms have been studied previously, its in-situ measurement of the motion of a PPy actuator was rarely investigated. To further expand the potential applications of a PPy actuator in automation and control of mesoscale robots, it is essential to develop a highly integrated sensor-actuator structure. In this work, we proposed a new fabrication process to make one kind by electrochemically synthesize a PPy layer on the opposite side of a carbon doped polydimethylsiloxane (CPDMS) based resistive sensor where a polyvinylidene fluoride (PVDF) membrane was used as a common substrate. A simple way to make a CPDMS sensor was proposed, and its electric and mechanical properties were studied using various material combinations. In addition, the characteristics of a CPDMSPPy composite was studied using cyclic voltammetry, and the integrated sensor response analysis with and without applied load during the electrochemically induced PPy actuation. The proposed structure was not only able to sense its bending motion, but it could also roughly distinguish the stiffness of an object in contact. To demonstrate a simple application, a compliant linkage structure, which simultaneously actuated and sensed by the proposed composite, was built.

Keywords: Modeling, Control, and Learning for Soft Robots, Soft Robot Materials and Design, Surgical Robotics: Steerable Catheters/Needles
Abstract: Soft continuum robots have the potential to revolutionize minimally invasive surgery. The challenges for such robots are universal; functioning within sensitive, unstructured and convoluted environments which are inconsistent between patients. As such, there exists an open design problem for robots of this genre. Research currently exists relating to the design considerations of on-board actuated soft robots such as fluid and tendon driven manipulators. Magnetically reactive robots, however, exhibit off-board actuation and consequently demonstrate far greater potential for miniaturization and dexterity. In this paper we present a soft, magnetically actuated, shape forming, high aspect ratio ‘tentacle-like’ robot. To overcome the aforementioned design challenges we also propose a novel design methodology based on a Neural Network trained using Finite Element Simulations. We demonstrate how our design approach generates static, two-dimensional tentacle profiles under homogeneous actuation based on predefined, desired deformations. To demonstrate our learnt approach, we fabricate and actuate candidate tentacles of 2mm diameter and 42mm length producing shape profiles within 5% mean absolute percentage error of simulations. With this proof of concept, we make the first step towards showing how tentacles with bespoke magnetic profiles may be designed and manufactured to suit specific anatomical constraints.

Keywords: Force and Tactile Sensing, Deep Learning in Robotics and Automation, Soft Sensors and Actuators
Abstract: In this paper, we introduce the novel soft tactile sensing system at large scale, which is developed based on vision and machine learning. The sensing platform, which was previous reported, is constructed from an elongated soft skin covering a bone-like transparent pipe. Two cameras are setup at top and bottom ends of the pipe for tracking movement of markers of the skin' inner side. By dividing the tactile sensing skin into 32 sub-areas, an indentor was controlled by a robot arm to push against each area at different contact depth (up to 24mm) for data collection. After that, a machine learning method was utilized to train the dataset to build a model for prediction of contact location and contact depth. We used the k-fold cross-validation theory to separate the dataset, then presented two networks to train the dataset; one is called Local Finding Network (LFD), and the other is Deformation Detecting Network (DDN), respectively. Base on these two networks, we obtained high accuracy results that can solve deficiencies caused by possible occlusions. The trained system was then demonstrated in realtime to show the potential of this system. The results in this research promises an efficient processing method toward accomplishment of large-scale tactile sensing systems.

Keywords: Soft Sensors and Actuators, Deep Learning in Robotics and Automation, Modeling, Control, and Learning for Soft Robots
Abstract: Soft strain sensors are becoming increasingly popular for obtaining tactile information in soft robotic applications. Diverse technological solutions are being investigated to design these sensors. Simultaneously, new methods for modeling these sensor are being proposed due to their highly nonlinear, time varying properties. Among them, machine learning based approaches, particularly using dynamic recurrent neural networks look the most promising. However, these complex networks have large number of free parameters to be tuned, making it difficult to apply them for real-world applications. This paper introduces the concept of transfer learning for modelling soft strain sensors, which allows us to utilize information learned in one task to be applied to another task. We demonstrate this technique on a passive anthropomorphic finger with embedded strain sensors used for two regression tasks. We show how the transfer learning approach can drastically reduce the number of free parameters to be tuned for learning new skills. This work is an important step towards scaling of sensor networks (algorithm-wise) and for using soft sensor data for high-level control tasks.

Keywords: Soft Robot Materials and Design, Biologically-Inspired Robots, Compliant Joint/Mechanism
Abstract: In nature, the softness, compliance and morphology of an organism’s body can generate efficient behaviours through interactions with the environment, without the need to involve the brain. Mimicking this approach in robotics could lead to more efficient systems that are easier to control. In this paper, we present a proof-of-concept for a bio-inspired soft robot that exploits its morphology and a very simple mechanical actuation method for locomotion. We draw inspiration from the jellyfish, one of the most efficient swimmer in our oceans, and design and fabricate soft tentacles based on the FinRay structure. We form the robot jellyfish by integrating four tentacles into a central mantle and actuate them through a rod coupled with a servo. Following initial tests, we improved the structure of the tentacles and the actuation control to generate a natural looking movement. The interaction with the surrounding environment was highlighted by the injection of fluorescent dye, imaged under UV. This showed the jetting behaviour of the robot jellyfish and the generation of vortices at the tip of the tentacles.

Keywords: Soft Robot Materials and Design, Cellular and Modular Robots, Compliant Joint/Mechanism
Abstract: Soft modular robots enable more flexibility and safer interaction with the changing environment than traditional robots. However, it has remained challenging to create deformable connectors that can be integrated into soft machines. In this work, we propose a flexible connector for soft modular robots based on micropatterned intersurface jamming. The connector is composed of micropatterned dry adhesives made by silicone rubber and a flexible main body with inflatable chambers for active engagement and disengagement. Through connection force tests, we evaluate the characteristics of the connector both in the linear direction and under rotational disruptions. The connector can stably support an average maximum load of 22 N (83 times the connector’s body weight) linearly and 10.86 N under planar rotation. The proposed connector demonstrates the potential to create a robust connection between soft modular robots without raising the system’s overall stiffness; thus guarantees high flexibility of the robotic system.

Keywords: Legged Robots, Biologically-Inspired Robots, Biomimetics
Abstract: Multi-legged animals (myriapods) such as centipedes move effectively in diverse terrain; flexible bodies and limbs allow them to morphologically adapt to the environment. To examine how the variation in body/limb forms of myriapods affects the mechanics and performance of terrestrial locomotion, we built a low-cost multi-legged hybrid (containing soft and hard components) robot which has 8 segments, each with two limbs driven out of phase. The back elements and limb pairs are driven by servo motors. Building on new theoretical results from geometric mechanics applied to myriapods, we systematically tested gait patterns with different leg contacts and body undulation on various laboratory and natural environments including flat and uneven rigid ground, stairs, and unstructured natural terrain (leaf litter, grass). On flat ground, the robot with rigid components moved in the same way as the theoretically predicted gaits. As the complexity of the environment increased, the robot's performance suffered (and theoretical predictions became unavailable) due to deleterious interactions like jamming of limbs. However, adding flexibility into the robot's body parts (legs, body joints, etc.) improved the open-loop locomotion performance (often to levels of that on flat ground) by either reducing the effects of environmental disturbances or increasing stability. Our findings show that in order to produce an agile, robust locomotive device, we need to understand the importance of body morphology and complex, dynamic interactions between an organism and its environment through systematic experiments in both the laboratory and natural environment.

Keywords: Modeling, Control, and Learning for Soft Robots, Soft Robot Materials and Design, Soft Robot Applications
Abstract: Continuum manipulator offers superior flexibility, deformability and adaptability to the environment, which makes it ideally suitable for safe interaction and for applications in confined spaces. Owning these advantages, the continuum manipulator gains increasing interests in the fields of surgical, underwater, inspection, etc. This paper introduces a novel continuum manipulator, which is designed as an independent and auxiliary modular for bringing extra dexterity and reachability to different rigid platforms. With the aim at precisely calculating the gripper pose of the manipulator, a probabilistic model-based approach is used, which learns a mapping among the actuator space and task space from experiments by using the dynamic mixture of Gaussians. The learned model and the control approach are validated with the help of a 3D trajectory tracking system. Finally, to test the versatility and reliability of the manipulator, we mount the actuation base of the continuum manipulator to a rigid manipulator, which are then used to manipulate an object in a confined space. The results of the experiments show that the proposed continuum manipulator can be controlled effectively by using a learned probabilistic model and the dexterity and workspace of a robotic system could be enhanced significantly by soft-rigid arm collaboration.

Keywords: Modeling, Control, and Learning for Soft Robots
Abstract: Given a configuration of a silicone soft robot, with the bounded installed actuators, this paper investigates the workspace estimation for a certain chosen points of interest for such a soft robot. For this, the Finite Element method is adopted to deduce the mathematical model of soft robots, based on which a forward-backward interval analysis approach is performed to estimate the workspace. Numerical simulations are provided to highlight the feasibility of the proposed methodology.

Keywords: Soft Robot Materials and Design, Soft Robot Applications, Compliant Joint/Mechanism
Abstract: Collision management strategies are an integral part of micro air vehicle (MAV) operation for flight sustainability. Among them, collision avoidance strategies require enhanced environmental and situational awareness for generating evasive maneuvers and collision-free trajectories. Simpler and more adaptable option is to prepare for collisions and design the physical system with predicted collision patterns in mind. In this work, a mechanically compliant quadcopter design using origami-inspired foldable robotics methods with protective shock absorbing elements has been proposed for a collision resilient quad-rotor MAV. 2D design of the foldable structure and the manufacturing process, including electronic hardware elements and software has been discussed. Our results show that in low speed collisions, the flight of the quadcopter is uninterrupted. The compliant quadcopter can continue flight after impact in near-hover conditions because of the reduction of impact forces due to the increased impact time.

Keywords: Soft Robot Applications, Soft Sensors and Actuators, Modeling, Control, and Learning for Soft Robots
Abstract: Affective touches play an important role in everyday communication because a significant amount of human interactions are through physical contacts. To facilitate robust, safe human-robot interactions (HRIs), we require robots with soft skins that can interpret affective touches. In this paper, we describe the fabrication of rapidly manufacturable, soft sensor skins using liquid metal embedded silicone elastomer as a resistive element and trained a recurrent neural network (RNN) to distinguish between a variety of pokes and rubs from human users. On a 2x2 sensor array, we obtained an average classification accuracy of 97% for ten different types of pokes and rubs, demonstrating that the combination of a soft sensor and machine learning can classify the interactions. Our approach is a step towards intelligent soft robots that can understand social interactions through touches.

Keywords: Biologically-Inspired Robots, Biomimetics, Modeling, Control, and Learning for Soft Robots
Abstract: In this study, we investigate whether an animal-in-the-loop (AIL) system can be applied to the collision avoidance ability of a bumblebee bombus ignitus. The AIL system is a novel experimental system for extracting the ability of animals to adapt to the environment and their own body by remotely controlling an autonomous mobile robot. The bumblebee causes an optomotor reflex to the optical flow and turns in the same yaw direction as it. However, it is unclear what kind of collision avoidance behavior is performed when flying toward a stationary obstacle. Therefore, we first measured collision avoidance behavior when flying toward a stationary obstacle. As a result, we found that the turning movement was elicited according to the expansion of the edge of the obstacle. Accordingly, we used this phenomenon for the AIL system in which the bumblebee remotely controls the quadcopter. As a result of presenting quadcopter first-person camera images to the bumblebee, we found avoidance behaviors elicited near the obstacle. This suggested that the AIL system can be applied to the collision avoidance ability of the bumblebee and that the behavior is changed by the expansion of the edge of the object.

Keywords: Sensor-based Control, Wearable Robots, Soft Sensors and Actuators
Abstract: Capacitive pressure sensors have several advantages in areas such as robotics, automation, aerospace, biomedical and consumer electronics. We present mathematical modelling, finite element analysis (FEA), fabrication and experimental characterization of ultra-low cost and paper-based touch mode flexible capacitive pressure sensor element using Do-It-Yourself (DIY) technology. The pressure sensing element is utilized to design large-area electronics skin for low-cost prosthetic hands. The presented sensor is characterize in normal, transition, touch and saturation modes. The sensor has higher sensitivity and linearity in touch mode operation from 10 to 40 kPa of applied pressure compared to the normal (0 to 8 kPa), transition (8 to 10 kPa) and saturation mode (after 40 kPa) with response time of 15.85 ms. Advantages of the presented sensor are higher sensitivity, linear response, less diaphragm area, less von Mises stress at the clamped edges region, low temperature drift, robust structure and less separation gap for large pressure measurement compared to normal mode capacitive pressure sensors. The linear range of pressure change is utilized for controlling the position of a servo motor for precise movement in robotic arm using wireless communication which can be utilized for designing skin-like structure for low-cost prosthetic hands.

Keywords: Soft Sensors and Actuators, Compliant Joint/Mechanism
Abstract: Twisted string actuators (TSAs) can produce linear motions by converting the electric motor's rotary motion. TSAs exhibit high power density, energy efficiency, and translational force. However, a common challenge to employ TSAs as soft and compliant actuators is to simultaneously increase their compliance and maximum strain. Previous studies predominantly utilized strings with high stiffness. Recent studies attempted to use strings with low stiffness with a sacrifice of maximum strain, especially under large loading conditions. In this paper, a novel strategy is proposed to simultaneously increase the compliance and maximum strain of TSAs. By replacing stiff strings with stretchable and coiled nylon strings, called supercoiled polymer (SCP) strings, the compliance of TSA is increased. Furthermore, by heating the TSA with SCP strings through Joule heating, additional linear contraction can be obtained. The fabrication procedure of the proposed TSAs is provided, and experimental characterizations of stiffness and actuation properties are conducted with different actuator configurations and loading conditions. Experiments confirm the enhanced performance of the proposed TSA using SCPs.

Keywords: Biologically-Inspired Robots, Distributed Robot Systems, Modeling, Control, and Learning for Soft Robots
Abstract: A peristaltic continuous mixing conveyor that focuses on the mechanism of the intestine has been developed as a technology for the mixing and transport of a solid–liquid mixture and high viscosity fluids. The developed peristaltic mixing conveyor succeeded in the mixing and transport of such mixtures in a previous study. A simple cyclic pattern is currently used as the movement pattern of the system, and the control method can be further improved. In this study, we aim to realize an intelligent mixing and generation of a conveyor pattern based on intestinal movement. The mixing state of content during activation was estimated through machine learning. The results from the time-series measurement data show that an internal mixing state in three unit types of the peristaltic mixing conveyor was successfully estimated.

Keywords: Soft Robot Applications, Grasping, Soft Sensors and Actuators
Abstract: Robotic hands with soft surfaces can perform stable grasping, but the high friction of the soft surfaces makes it difficult to release objects, or to perform operations that require sliding. To solve this issue, we previously developed a contact area variable surface (CAVS), whose friction changed according to the load. However, only our fundamental results were previously presented, with detailed analyses not provided. In this study, we first investigated the CAVS friction anisotropy, and demonstrated that the longitudinal direction exhibited a larger ratio of friction change. Next, we proposed a ‘sensible’ CAVS, capable of providing a variable-friction mechanism, and tested its sensing and control systems in operations requiring switching between sliding and stable-grasping modes. Friction sensing was performed using an embedded camera, and we developed a gripper using the sensible CAVS, considering the CAVS friction anisotropy. In CAVS, the low-friction mode corresponds to a small grasping force, while the high-friction mode corresponds to a greater grasping force. Therefore, by controlling only the friction mode, the gripper mode can be set to either the sliding or stable-grasping mode. Based on this feature, a methodology for controlling the contact mode was constructed. We demonstrated a manipulation involving sliding and stable grasping, and thus verified the efficacy of the developed sensible CAVS.

Keywords: Biologically-Inspired Robots, Soft Robot Materials and Design, Hydraulic/Pneumatic Actuators
Abstract: Soil exploration is required in many different activities, ranging from detecting minerals or pollutants to finding water sources. Yet, penetrating soil is extremely challenging because of high pressures and friction even in a few centimeters of depth. In nature, earthworms move inside soils by peristalsis, i.e. by wave-like deformation traveling along their entire soft body. Taking inspiration from these animals and understanding their peristaltic locomotion offer interesting solutions for achieving soil exploration. To this aim, we fabricated and assembled a soft robot able to generate controlled and tunable peristaltic deformations. We systematically experimented peristaltic locomotion in a granular medium and compared it to movement on a planar surface. Our results show that locomotion based on peristaltic waves of a soft robot in granular media is possible, generally slower than locomotion on planar surfaces. This helps in reducing friction during the robot penetration in soil. Interestingly, our results also show how the wave patterns and actuation frequency differently influence the speed and the direction of movement in the different media, providing guidance in developing future soft robots for soil exploration.

Keywords: Soft Sensors and Actuators, Deep Learning in Robotics and Automation, Haptics and Haptic Interfaces
Abstract: Bubble arrays are new type of devices that can render forces to the user's skin by pressurizing thin membranes. These devices are suitable for wearables due to being soft while having low encumbrance to the user body. In this manuscript, we describe a solution to tackle the major challenge towards controllability of this devices: state estimation of the membrane. This work describes the manufacturing of a self-sensing composite elastomeric membrane, device integration and the methods for state estimation using deep learning architectures. Additive manufacturing was used to fabricate the membrane and a pneumatic bubble actuator was operated at low pressures (<10 psi). This membrane is thin (~700um), soft (∼3 MPa), and includes a matrix of elastomeric capacitors for distributed sensing of deformation. Our model predicted the magnitude of deformation with a mean absolute error MAE= 0.321mm corresponding to 13.1% accuracy.

Keywords: Soft Sensors and Actuators, Soft Robot Materials and Design, Hydraulic/Pneumatic Actuators
Abstract: Untethered control of soft-bodied robots is attractive for interactions in a variety of unstructured and dynamic environments. However, soft robotics systems are currently limited in terms of wireless, selective, and scalable control of multiple actuators. Therefore, we propose a method to wirelessly drive multiple soft actuators by laser projection. A small amount of low-boiling-point liquid inside a planar thin pouch can be heated by a laser and evaporated to inflate the whole body. Laser projection enables both wireless energy supply and the selection of target actuators. Further, the low-boiling-point liquid serves as an actuation source and as a receiver of laser irradiation. Thus, we do not need additional components such as electric circuits and batteries to achieve simple and scalable implementation of multiple soft actuators. We evaluated the mechanical properties and demonstrated that the system can wirelessly control the gestures of fingers of a robot hand. We also verified that our method can activate a group of mobile soft robots simultaneously and individually while tracking the actuator positions. Our approach contributes to the scalable deployment of soft robotic systems by removing tethers for power and communication.

Keywords: Soft Sensors and Actuators, Soft Robot Materials and Design, Soft Robot Applications
Abstract: An ionic polymer metal composite (IPMC) actuator is one of the most promising electroactive polymer devices. It is a lightweight and flexible actuator that can be operated in low voltage, high-frequency bandwidth, and even water. To widen the application field and enhance the potential of the IPMC actuator, a stable and flexible arbitrary 3D fabrication method is required. To address this arbitrary 3D fabrication problem, we have proposed and studied a novel fabrication method using papers and fabrics as the base materials of IPMC actuators. In this paper, we propose a novel paper/fabric (PF)-IPMC and its detailed fabrication process. For future design optimization, the fundamental characteristics derived from a type of paper and fabric are also determined through experiments. PF-IPMCs created from papers (Kimwipe and Kimtowel) have a greater ability for deformation, blocking force, and adsorption ratio than those created from fabrics (cotton and polyester). After the fundamental properties were evaluated, an origami plane and a crane PF-IPMC robots were created using the proposed method. Both PF-IPMC robots retained their initial shapes after completion of the fabrication process and could be operated normally. This shows the potential of the PF-IPMC for application to 3D craft.

Keywords: Grippers and Other End-Effectors, Grasping, Soft Robot Applications
Abstract: This paper proposes a procedure for the rapid prototyping and on-demand manufacturing of thin film flexible electro-adhesive devices (EADs) made with a commercial polyimide dielectric layer, inkjet printed interdigitated silver electrodes and blade coated silicone elastomer encapsulation backing. As a proof demonstration, flexible thin-film EADs featuring 9.6 cm² active area, 315 µm thickness and 0.7 g weight have been manufactured and tested over different adhering substrates showing peak adhesive shear stresses of up to 56.67 kPa, fast response time (11 ms for initial activation and 0.3 s for full electrification) and little energy requirements (from 1.3 mJ for initial activation to 20 mJ for full electrification and with a subsequent power consumption of about 1 mW for long-term grasp holding). Practical application of the manufactured EADs within a gripper for the grasping and handling of real objects that include a glass bottle, a hollow carbon fiber tube, a cardboard box, a box with thin polypropylene envelope and a polypropylene bottle is also demonstrated.

Keywords: Soft Robot Materials and Design
Abstract: Functional, responsive materials are attractive for use as key components in soft robots as they can replace otherwise rigid or bulky parts. In this work, we present a functional silicone with shape-memory properties that, due to the retention of elasticity and flexibility, can be seamlessly integrated into the body of a soft robot. By dispersing particles of low-melting-point metal alloy (Field's metal) into a silicone matrix, the resulting composite can be ``frozen'' into various shapes by sequentially heating to melt the Field's metal particles, stretching the composite, and cooling to solidify the Field's metal particles in the deformed configuration. The ramifications of this operational capability include both stiffness control and 3D shape reconfiguration. In this paper, we characterize the thermomechanical behavior and shape memory performance of the Field's metal/silicone composite. We then highlight applications of the material to impedance and trajectory control, and topology recording.

Keywords: Soft Sensors and Actuators, Soft Robot Materials and Design, Wearable Robots
Abstract: Soft ionic sensing represents a promising technology to develop low-cost, biocompatible sensors that can be used to provide exteroception and interoception capabilities to soft robotic devices, or in wearable applications to measure joint motion and forces on the body. In this paper, we present the design, fabrication, and testing of an ionic soft sensor based on a NaCl solution that can measure pressure and strain independently. The proposed system uses a singular output to simultaneously measure strain and pressure via the modification of the input signal frequency and amplitude, effectively decoupling the detection of forces along different axes and allowing the user to track these inputs in real-time.

Keywords: Force and Tactile Sensing, Perception for Grasping and Manipulation, Learning and Adaptive Systems
Abstract: In object picking, knowing the state of picked up objects is very important to conduct succeeding tasks surely. Especially, the ability to count objects in hand is crucial to judge whether the previous picking action was successful or not. This work seeks to endow such ability to a robot manipulator. Vision-based methods cannot be relied on to count objects in hand due to the occlusion problem, especially when dealing with objects smaller than one centimeter like small screws. Hence, number estimation should be conducted from tactile sensor information. However, compact pressure-based tactile sensor arrays can not take fine outlines of such small objects because sensor element size is not small enough, meaning that a simple rule-based approach is not feasible. Furthermore, the tactile sensor array fixed on a rigid plane surface can only contact protruding parts of in-hand objects; thus, the simple installation of tactile sensor arrays on a manipulator surface is insufficient for counting objects. Therefore, in this work, we propose 1)a number estimation method which uses a convolution neural network and 2) to cover a tactile sensor array with soft material to enrich tactile information to improve estimation accuracy. We validated the proposed method using data collected by a simple gripper and achieved 89% accuracy in estimating the small screw number in the gripper.

Keywords: Soft Robot Materials and Design, Soft Sensors and Actuators, Soft Robot Applications
Abstract: Robotics is a key technology to improve and support the medical services by increasing efficiency and improving quality of care. One particular challenge is medical palpation, where a physicians uses their hands to manually feel and identify abnormalities or tumors. This is a challenging procedure that takes many years to learn. A soft robotic phantom which could provide feedback on palpation technique could enable medical professionals to learn more easily. Furthermore, this would allow quantitative information about how practitioners perform the technique to be gained. This can be used to inform and move towards robotic palpation that mirrors the human performance. In this work, we demonstrate an approach for developing a large-area sensorized phantom that uses sensor morphology to sense over a large areas. A reconstruction algorithm allows the depth of palpation to be measured and the location to be identified with a precision of 5mm. This phantom was used to measure deformation from human palpation, with the resultant information used to replicate the behaviour with a robot hand. To the authors’ knowledge this is the first sensorized phantom that characterizes large scale deformation from human palpation to inform the performance of the procedure by a robot.

Keywords: Soft Sensors and Actuators, Soft Robot Applications, Soft Robot Materials and Design
Abstract: Shape reconstruction by soft sensors may be useful in applications ranging from precision agriculture to haptics and factory automation due to the potential for low-cost fabrication, durable operation, and safe and compliant interaction. Current prevalent techniques, however, require expertise, expensive materials, and high-end processing equipment which limits both their transition to practice and their accessibility to researchers. To address this issue, we present easily accessible, low-cost, and rapid fabrication techniques for soft and resistive carbon composite sensors. We characterize their repeatability and durability in response to stretch up to 135%. We further show how this fabrication technique may be easily customized to two different applications, including a stretchable, tactile interface for passive sensing, and an active, soft pneumatic gripper that can fully encompass an object to sense its shape. We complement these with simple control and analysis, and show how to achieve high relative accuracy, despite the high manufacturing tolerances of the sensors.

Keywords: Soft Robot Materials and Design, Soft Robot Applications, Grasping
Abstract: This paper describes a new type of compliant and configurable soft robot, a boundary-constrained swarm. The robot consists of a sealed flexible membrane that constrains both a number of mobile robotic subunits and passive granular material. The robot can change the volume fraction of the sealed membrane by applying a vacuum, which gives the robot the ability to operate in two distinct states: compliant and jammed. The compliant state allows the robot to surround and conform to objects or pass through narrow corridors. Jamming allows the robot to form a desired shape; grasp, manipulate, and exert relatively high forces on external objects; and achieve relatively higher locomotion speeds. Locomotion is achieved with a combination of whegs (wheeled legs) and vibration motors that are located on the robotic subunits. The paper describes the mechanical design of the robot, the control methodology, and its object handling capability.

Keywords: Biologically-Inspired Robots, Soft Robot Materials and Design, Soft Robot Applications
Abstract: In this paper we propose a novel pneumatic soft snake robot which exploits traveling-wave motion to move in complex, constrained environments such as a pipeline. The robot is modular, with a unique pneumatic system design that requires the use of only four air channels regardless of the number of modules. The robot is 3D-printed, and thus low-cost and easy to build. Finite element modeling of the bending behavior of each module is conducted in ANSYS. The dynamic behavior of the robot, consisting of six modules, is further modeled in SOFA. In particular, it is found that the locomotion speed of the robot increases with the actuation pressure and decreases with the friction coefficient. Extensive experimental results on a snake robot prototype show good agreement with model predictions. The robot also demonstrates the capability of moving in constrained pipeline environments, including travelling in pipes of different diameters and challenging geometry such as a sharp elbow.

Keywords: Soft Robot Materials and Design
Abstract: Soft continuum manipulators provide a safe alternative to traditional rigid manipulators, because their bodies can absorb and distribute contact forces. Soft manipulators have near infinite potential degrees of freedom, but a limited number of control inputs. This underactuation means soft continuum manipulators often lack either the controllability or the dexterity to achieve desired tasks. In this work, we present an extension of McKibben actuators, which have well-known models, that increases the controllable degrees of freedom using active reconfiguration of the constraining fibers. These Active Fiber Reinforced Elastomeric Enclosures (AFREEs) preform some combination of length change and twisting, depending on the fiber configuration. Experimental results shows that by changing the fiber angles within a range of -30 to 30 degrees and actuating the resulting configuration between 10.3 kPa and 24.1 kPa, we can achieve twists between +/-60 degrees and displacements between -2 and 4 mm. By additionally controlling the fiber lengths and pressure, we can modify the AFREE kinematics further, creating dynamic behaviors and trajectories of actuation. The presented actuator creates the possibility to reconfigure actuator kinematics to meet desired soft robot motions.

Keywords: Soft Robot Materials and Design, Biomimetics, Tendon/Wire Mechanism
Abstract: In this paper, we developed, for humanoid robots, a lightweight spine mechanism that deforms flexibly, drawing continuous curves. This mechanism is the one that mimics the human musculoskeletal structure. Although the human spine has a high degree of freedom and is difficult to control, our mechanism solves the degree of freedom problem with a tendondriven underdrive mechanism that applies the technique of continuum robots, and with only nine motors, the basic operations of flexion, extension, lateral flexion and rotation are all realized for each of the three curves, i.e., the cervical spine, thoracic spine, and lumbar spine. A slender and lightweight mechanism combining coil springs and CF (Carbon Fiber) rods is driven by pulling 24 threads, and these threads are allocated to a total of 9 motors. The reproducibility of the basic motions of the human spine, such as flexion, extension, lateral flexion and rotation, was shown by actual machine tests. As an example, we reproduced ‘arabesque’ which is one of the typical ballet postures.

Keywords: Soft Robot Materials and Design
Abstract: Soft, stretchable sensors, such as artificial skins or tactile sensors, are attractive for numerous soft robotic applications due to the low material compliance. Conductive polymers are a necessary component of many soft sensors, and this work presents the electromechanical characterization of 3D-printable conductive polymer composites. Dog-bone shaped samples were 3D printed using a digital light processing (DLP)-based 3D printer for characterization. The 3D printable resin consists of monomer, crosslinker, conductive nano-filler, and a photo-initiator. The characterization was performed in two tracks. First, the effect of two different crosslinkers was investigated with different compositions and second, the effect of concentration of conductive nano-fillers was explored. Crosslinkers were chosen by referring to previous studies, and carbon nanotubes (CNTs) were utilized as conductive nano-fillers. The samples were 3D printed and characterized using an electromechanical test setup. To demonstrate utility for 3D printed soft robotics, a capacitance-based joystick sensor composed of both conductive and non-conductive resins was 3D printed.

Keywords: Soft Sensors and Actuators, Human Factors and Human-in-the-Loop, Soft Robot Applications
Abstract: Conventional planar pneumatic actuators have not been able to make the most of their thin, lightweight, and flexible nature due to tubes and pumps inevitably connected to them. In this paper, we build upon our prior work and propose Liquid Pouch Motors, a family of printable soft actuators that consist of one or more gas-tight bladders (called pouches) filled with low boiling point liquid. When the heat over 34 °C is applied to them, the liquid inside the pouch evaporates and the whole structure inflates. Especially, we newly presented roll-to-roll mass production of Liquid Pouch Motors with the rectangular form factor in addition to the previously reported CNC heat drawing. We also discussed the suitable material selection and the volume of liquid needed for the actuators. Based on this fabrication procedure, we evaluated the mechanical properties of the proposed actuators. By leveraging the benefits of the actuators, we then implemented three shape-changing interfaces: (1) electrically driven printable paper robots combined with printed paper circuit; (2) an untethered architecture facade that passively actuates by ambient heat; and (3) dresses that change their color and texture by the incident light and the body temperature. We believe that Liquid Pouch Motors will work as a new toolbox for the fabrication and application of soft actuators in interactive scenarios.

Keywords: Force and Tactile Sensing, Soft Sensors and Actuators, Deep Learning in Robotics and Automation
Abstract: This paper describes the design of a multi-camera optical tactile sensor that provides information about the contact force distribution applied to its soft surface. This information is contained in the motion of spherical particles spread within the surface, which deforms when subject to force. The small embedded cameras capture images of the different particle patterns that are then mapped to the three-dimensional contact force distribution through a machine learning architecture. The design proposed in this paper exhibits a larger contact surface and a thinner structure than most of the existing camera-based tactile sensors, without the use of additional reflecting components such as mirrors. A modular implementation of the learning architecture is discussed that facilitates the scalability to larger surfaces such as robotic skins.

Keywords: Soft Sensors and Actuators
Abstract: The sensor morphology is of great importance, especially for tactile sensors that mainly transduce physical contacts to perceivable signals for further processing by the central system. Sensors with various morphology can result in different signal thus potentially influencing perception. Most previous researches on tactile sensor morphology focused on revealing advance of a particular morphology in a binary manner, by providing sensors with and without that morphology then comparing the performance. Less attention was paid to sensors with changeable morphology. This letter demonstrates a wrinkled soft tactile sensor with variable morphology, leading to tunable sensing and perception property. Specifically, the geometrical structure of wrinkles and the overall stiffness of the sensor can be varied by changing the sensor bending state. This morphing behavior was modeled analytically then verified by real experiment. Furthermore, these morphological variations were found to closely affect the performance of sensor in three tasks including force sensing, shape discrimination and texture detection. Interestingly, for each task, there was always an optimal morphological state that maximized the sensor performance. This sheds light to novel active tactile sensing system design, thus realizing "active" by adapting morphology instead of sensorimotor control algorithm to optimize perception gain.

Keywords: Modeling, Control, and Learning for Soft Robots, Hydraulic/Pneumatic Actuators, Optimization and Optimal Control
Abstract: Soft robotics holds enormous promise for a wide class of applications. However, system controllability, bandwidth, portability, and energy efficiency of soft robot power supplies are often inadequate. Soft robotics desperately needs improved solutions to drive soft actuators either pneumatically or hydraulically. This research paper offers a contribution to bridge this gap. It deals with small-scale power supplies for hydraulically-driven soft robots based on fluidic elastomer actuators in the power range 5-400 W. A design procedure for such power supplies is developed with an emphasis on high-bandwidth control. The performance requirements are established based on a literature survey, and design alternatives are discussed. Then, a prototype consisting of a 0.102 cm3/rev positive-displacement, external gear pump directly driven by a 70 W brushless dc motor is built and experimentally characterized. Control of the targeted soft robots results possible because the proposed power supply has high-bandwidth varying between about 12 and 30 Hz (i.e., well above the actuators’ mechanical frequency). The power supply’s energy efficiency is also measured, and areas for its improvement are identified. The proposed solution is feasible, applicable to autonomous soft robots, scalable to the power levels of interest, and possibly useful for both tethered soft robots and rigid/semi-rigid robots.

Keywords: Soft Robot Materials and Design, Additive Manufacturing, Soft Robot Applications
Abstract: Recent advances in flexible electronics, soft sensors and soft actuators are paving the way towards replacing hard printed circuit boards for soft counterparts in various applications (e.g. soft robotics, wearable devices, etc.). The need to achieve robust electrical connections between both soft components and traditional rigid components poses many challenges. In particular, the in-extensiblility of commercially available interconnects (e.g. single/multi-strand conductive wires, conductive metallic tapes, etc.) can affect the structural properties of soft components. Herein we present the design and demonstrate the fabrication method for making flexible fiber interconnects (FFI) by printing flexible guidepaths and simultaneously layering and embedding conductive yarns within. The effectiveness and robustness of the flexible interconnects for use within soft structures is characterized. Simple FFI designs can be used within structures undergoing up to 200% strains without interfering with the substrate stress-strain behavior. Electrical conductivity is also shown to be stable even during cyclic loading.

Keywords: Soft Robot Materials and Design, Simulation and Animation, Modeling, Control, and Learning for Soft Robots
Abstract: The manual design of soft robots and their controllers is notoriously challenging, but it could be augmented -- or, in some cases, entirely replaced -- by automated design tools. Machine learning algorithms can automatically propose, test, and refine designs in simulation, and the most promising ones can then be manufactured in reality (sim2real). However, it is currently not known how to guarantee that behavior generated in simulation can be preserved when deployed in reality. Although many previous studies have devised training protocols that facilitate sim2real transfer of control polices, little to no work has investigated the simulation-reality gap as a function of morphology. This is due in part to an overall lack of tools capable of systematically designing and rapidly manufacturing robots. Here we introduce a low cost, open source, and modular soft robot design and construction kit, and use it to simulate, fabricate, and measure the simulation-reality gap of minimally complex yet soft, locomoting machines. We prove the scalability of this approach by transferring an order of magnitude more robot designs from simulation to reality than any other method. The kit and its instructions can be found here: https://github.com/skriegman/sim2real4designs

Keywords: Soft Sensors and Actuators, Soft Robot Materials and Design, Soft Robot Applications
Abstract: Shape memory alloy (SMA) has been widely used in soft robotics systems due to its high work density, shape programmability, rigidity tunability and low requirement of the peripheral electronic devices and power for control and actuation. However, the low bandwith of the SMA actuator resulted in long cooling time that limit the performance of the soft robotics. To address this issue, methods like embedding thermally conductive elastomer and adding antagonistic mechanisms have been attempted. Here, we combine both methods and construct a series of study to characterize the improvement in actuation performance of the SMA actuators both in air and water by embedding the SMA wire inside liquid metal embedded elastomer with various liquid metal volume ratio. The improvement in actuation frequency leads to a locomotion speed improvement of a crawling and frog-inspired swimming soft robot.

Keywords: Legged Robots, Soft Robot Applications, Tendon/Wire Mechanism
Abstract: We report on experiments with a laptop-sized (0.23m, 2.53kg), paper origami robot that exhibits highly dynamic and stable two degree-of-freedom (circular boom) hopping at speeds in excess of 1.5 bl/s (body-lengths per second) at a specific resistance O(1) while achieving aerial phase apex states 25% above the stance height over thousands of cycles. Three conventional brushless DC motors load energy into the folded paper springs through pulley-borne cables whose sudden loss of tension upon touchdown triggers the release of spring potential that accelerates the body back through liftoff to flight with a 20W powerstroke, whereupon the toe angle is adjusted to regulate fore-aft speed. We also demonstrate in the vertical hopping mode the transparency of this actuation scheme by using proprioceptive contact detection with only motor encoder sensing. The combination of actuation and sensing shows potential to lower system complexity for tendon-driven robots.

Keywords: Biologically-Inspired Robots, Compliant Joint/Mechanism, Legged Robots
Abstract: Recent studies on dynamic legged locomotion have focused on incorporating passive compliant elements into robot legs which can help with energy efficiency and stability, enabling them to work in wide range of environments. In this work, we present the design and testing of a soft robotic foot capable of active stiffness control using granular jamming. This foot is designed and tested to be used on soft, flowable ground such as sand. Granular jamming feet enable passive foot shape change when in contact with the ground for adaptability to uneven surfaces, and can also actively change stiffness for the ability to apply sufficient propulsion forces. We seek to study the role of shape change and stiffness change in foot-ground interactions during foot-fall impact and shear. We have measured the acceleration during impact, surface traction force, and the force to pull the foot out of the medium for different states of the foot. We have demonstrated that the control of foot stiffness and shape using the proposed foot design leads to improved locomotion, specifically a ~52 % reduced foot deceleration at the joints after impact, ~63 % reduced depth of penetration in the sand on impact, higher shear force capabilities for a constant depth above the ground, and ~98 % reduced pullout force compared to a rigid foot.

Keywords: Soft Robot Materials and Design, Legged Robots, Soft Robot Applications
Abstract: Walking on natural terrain like soil and rock is a challenging problem that has been approached from a variety of strategies such as using sophisticated control methods, compliant legs, and compliant feet. In this paper we explore how to modify granular jamming feet for walking applications by adding stabilizing internal structures. Previous work has explored how granular jamming technology can be used to create compliant and stiffness changing feet that enable locomotion over a diverse range of natural terrain by allowing robot feet to conform around 3D multicomponent terrain such as wood chips and gravel and stiffen, preventing slip. To date, no work has been done to tune granular jamming feet for the specific application of walking. We show that adding internal structures to granular jamming membranes can increase the force they are able to resist without slipping by 1.5x while maintaining their ability to conform around obstacles. When attached to a robot, we see increases in speed of up to 1.4x, decreases in the duty cycle necessary to reach desired foot trajectories of up to 5%, and increases in traction force of up to 1.2x over a diverse set of natural terrain.

Keywords: Soft Robot Materials and Design, Legged Robots, Biologically-Inspired Robots
Abstract: Spinal-driven locomotion was first hypothesized to exist in biological systems in the 1980's; however, only recently has the concept been applied to legged robots. In implementing spinal-driven locomotion in robots to-date, researchers have focused on bending in the spine. In this paper, we propose an additional mode of spinal-driven locomotion: axial torsion via helical actuation patterns. To study torsional spinal-driven locomotion, a six-legged robot with unactuated legs is used. This robot is designed to be modular to allow for changes in the physical system, such as material stiffness of the spine and legs, and has actuators that spiral around the central elastomeric spine of the robot. A model is provided to explain torsional spinal-driven locomotion. Three spinal gaits are developed to allow the robot to walk forward, through which we demonstrate that the speed of the robot can be influenced by the stiffness of the spine and legs. We also demonstrate that a single gait can be used to drive the robot forward and turn the robot left and right by adjusting the leg positions or foot friction. The results indicate that the inclusion of helical actuation patterns can assist in movement. The addition of these actuation patterns or active axial torsion to future, more complex robots with active leg control may enhance the energy efficiency of locomotion or enable fast, dynamic maneuvering.