Exploiting the Dynamics of Soft Materials for Machine Learning

Embedding Bifurcations into Pneumatic Artificial Muscle

Harnessing complex body dynamics has long been a challenge in robotics, particularly when dealing with soft dynamics, which exhibit high complexity in interacting with the environment. Recent studies indicate that these dynamics can be used as a computational resource, exemplified by the McKibben pneumatic artificial muscle, a common soft actuator. This study demonstrates that bifurcations, including periodic and chaotic dynamics, can be embedded into the pneumatic artificial muscle, with the entire bifurcation structure using the framework of physical reservoir computing. These results suggest that dynamics not present in training data can be embedded through bifurcation embedment, implying the capability to incorporate various qualitatively different patterns into pneumatic artificial muscle without the need to design and learn all required patterns explicitly. Thus, this study introduces a novel approach to simplify robotic devices and control training by reducing reliance on external pattern generators and the amount and types of training data needed for control.

Soft Sensory-Motor System Based on Ionic Solution for Robotic Applications

Soft robots claim the architecture of actuators, sensors, and computation demands with their soft bodies by obtaining fast responses and adapting to the environment. Sensory-motor coordination is one of the main design principles utilized for soft robots because it allows the capability to sense and actuate mutually in the environment, thereby achieving rapid response performance. This work intends to study the response for a system that presents coupled actuation and sensing functions simultaneously and is integrated in an arbitrary elastic structure with ionic conduction elements, called as soft sensory-motor system based on ionic solution (SSMS-IS). This study provides a comparative analysis of the performance of SSMS-IS prototypes with three diverse designs: toroidal, semi-toroidal, and rectangular geometries, based on a series of performance experiments, such as sensitivity, drift, and durability. The design with the best performance was the rectangular SSMS-IS using silicon rubber RPRO20 for both internal and external pressures applied in the system. Moreover, this work explores the performance of a bioinspired soft robot using rectangular SSMS-IS elements integrated in its body. Further, it investigated the feasibility of the robot to adapt its morphology online for environment variability, responding to external stimuli from the environment with different levels of stiffness and damping.

Embodying Multifunctional Mechano-Intelligence in and Through Phononic Metastructures Harnessing Physical Reservoir Computing

Recent advances in autonomous systems have prompted a strong demand for the next generation of adaptive structures and materials to possess built-in intelligence in their mechanical domain, the so-called mechano-intelligence (MI). Previous MI attempts mainly focused on specific case studies and lacked a systematic foundation in effectively and efficiently constructing and integrating different intelligent functions. Here, a new approach is uncovered to create multifunctional MI in adaptive structures using physical reservoir computing (PRC). That is, to concurrently embody computing power and the key elements of intelligence, namely perception, decision-making, and commanding, directly in the mechanical domain, advancing from conventional reliance on add-on computers and massive electronics. As an exemplar platform, a mechanically intelligent phononic metastructure is developed by harnessing its high-degree-of-freedom nonlinear dynamics as PRC power. Through analyses and experiments, multiple intelligent structural functions are demonstrated ranging from self-tuning wave controls to wave-based logic gates. This research provides the much-needed basis for creating future smart structures and materials that greatly surpass the state of the art-such as lower power consumption, more direct interactions, and better survivability in harsh environments or under cyberattacks. Moreover, it enables the addition of new functions and autonomy to systems without overburdening the onboard computers.

Handwritten digit recognition by spin waves in a Skyrmion reservoir

By performing numerical simulations for the handwritten digit recognition task, we demonstrate that a magnetic skyrmion lattice confined in a thin-plate magnet possesses high capability of reservoir computing. We obtain a high recognition rate of more than 88%, higher by about 10% than a baseline taken as the echo state network model. We find that this excellent performance arises from enhanced nonlinearity in the transformation which maps the input data onto an information space with higher dimensions, carried by interferences of spin waves in the skyrmion lattice. Because the skyrmions require only application of static magnetic field instead of nanofabrication for their creation in contrast to other spintronics reservoirs, our result consolidates the high potential of skyrmions for application to reservoir computing devices.

Characterization of Information-Transmitting Materials Produced in Ionic Liquid-based Neuromorphic Electrochemical Devices for Physical Reservoir Computing

Device implementation of reservoir computing, which is expected to enable high-performance data processing in simple neural networks at a low computational cost, is an important technology to accelerate the use of artificial intelligence in the real-world edge computing domain. Here, we propose an ionic liquid-based physical reservoir device (IL-PRD), in which copper cations dissolved in an IL induce diverse electrochemical current responses. The origin of the electrochemical current from the IL-PRD was investigated spectroscopically in detail. After operating the device under various operating conditions, X-ray photoelectron spectroscopy of the IL-PRD revealed that electrochemical reactions involving Cu, Cu2O, Cu(OH)2, CuSx, and H2O occur at the Pt electrode/IL interface. These products are considered information transmission materials in IL-PRD similar to neurotransmitters in biological neurons. By introducing the Faradaic current components due to the electrochemical reactions of these materials into the output signal of IL-PRD, we succeeded in improving the time-series data processing performance of the nonlinear autoregressive moving average task. In addition, the information processing efficiency in machine learning to classify electrocardiogram signal waveforms was successfully improved by using the output current from IL-PRD. Optimizing the electrochemical reaction products of IL-PRD is expected to advance data processing technology in society.

Leveraging plant physiological dynamics using physical reservoir computing

Plants are complex organisms subject to variable environmental conditions, which influence their physiology and phenotype dynamically. We propose to interpret plants as reservoirs in physical reservoir computing. The physical reservoir computing paradigm originates from computer science; instead of relying on Boolean circuits to perform computations, any substrate that exhibits complex non-linear and temporal dynamics can serve as a computing element. Here, we present the first application of physical reservoir computing with plants. In addition to investigating classical benchmark tasks, we show that Fragaria × ananassa (strawberry) plants can solve environmental and eco-physiological tasks using only eight leaf thickness sensors. Although the results indicate that plants are not suitable for general-purpose computation but are well-suited for eco-physiological tasks such as photosynthetic rate and transpiration rate. Having the means to investigate the information processing by plants improves quantification and understanding of integrative plant responses to dynamic changes in their environment. This first demonstration of physical reservoir computing with plants is key for transitioning towards a holistic view of phenotyping and early stress detection in precision agriculture applications since physical reservoir computing enables us to analyse plant responses in a general way: environmental changes are processed by plants to optimise their phenotype.

Recent advances in biomimetic soft robotics: fabrication approaches, driven strategies and applications

Compared to traditional rigid-bodied robots, soft robots are constructed using physically flexible/elastic bodies and electronics to mimic nature and enable novel applications in industry, healthcare, aviation, military, etc. Recently, the fabrication of robots on soft matter with great flexibility and compliance has enabled smooth and sophisticated 'multi-degree-of-freedom' 3D actuation to seamlessly interact with humans, other organisms and non-idealized environments in a highly complex and controllable manner. Herein, we summarize the fabrication approaches, driving strategies, novel applications, and future trends of soft robots. Firstly, we introduce the different fabrication approaches to prepare soft robots and compare and systematically discuss their advantages and disadvantages. Then, we present the actuator-based and material-based driving strategies of soft robotics and their characteristics. The representative applications of soft robotics in artificial intelligence, medicine, sensors, and engineering are summarized. Also, some remaining challenges and future perspectives in soft robotics are provided. This work highlights the recent advances of soft robotics in terms of functional material selection, structure design, control strategies and biomimicry, providing useful insights into the development of next-generation functional soft robotics.

Dynamic Nonlinear Behavior of Ionic Liquid-Based Reservoir Computing Devices

Herein, a physical reservoir device that uses faradaic currents generated by redox reactions of metal ions in ionic liquids was developed. Synthetic time-series data consisting of randomly arranged binary number sequences ("1" and "0") were applied as isosceles-triangular voltage pulses with positive and negative voltage heights, respectively, and the effects of the faradaic current on short-term memory and parity-check task accuracies were verified. The current signal for the first half of the triangular voltage-pulse period, which contained a much higher faradaic current component compared to that of the second half of the triangular voltage-pulse period, enabled higher short-term memory task accuracy. Furthermore, when parity-check tasks were performed using a faradaic current generated by asymmetric triangular voltage-pulse levels of 1 and 0, the parity-check task accuracy was approximately eight times higher than that of the symmetric triangular voltage pulse in terms of the correlation coefficient between the output signal and target data. These results demonstrate the advantage of the faradaic current on both the short-term memory characteristics and nonlinear conversion capabilities and are expected to provide guidance for designing and controlling various physical reservoir devices that utilize electrochemical reactions.

An Embodied Intelligence-Based Biologically Inspired Strategy for Searching a Moving Target

Bacterial chemotaxis in unicellular Escherichia coli, the simplest biological creature, enables it to perform effective searching behaviour even with a single sensor, achieved via a sequence of "tumbling" and "swimming" behaviours guided by gradient information. Recent studies show that suitable random walk strategies may guide the behaviour in the absence of gradient information. This article presents a novel and minimalistic biologically inspired search strategy inspired by bacterial chemotaxis and embodied intelligence concept: a concept stating that intelligent behaviour is a result of the interaction among the "brain," body morphology including the sensory sensitivity tuned by the morphology, and the environment. Specifically, we present bacterial chemotaxis inspired searching behaviour with and without gradient information based on biological fluctuation framework: a mathematical framework that explains how biological creatures utilize noises in their behaviour. Via extensive simulation of a single sensor mobile robot that searches for a moving target, we will demonstrate how the effectiveness of the search depends on the sensory sensitivity and the inherent random walk strategies produced by the brain of the robot, comprising Ballistic, Levy, Brownian, and Stationary search. The result demonstrates the importance of embodied intelligence even in a behaviour inspired by the simplest creature.

A review of motor neural system robotic modeling approaches and instruments

In this review, we are considering an actively developing tool in neuroscience-robotic modeling. The new perspective and existing application fields, tools, and methods are discussed. We try to determine starting positions and approaches that are useful at the beginning of new research in this field. Among multiple directions of the research is robotic modeling on the level of muscles fibers and their afferents, skin surface sensors, muscles, and joints proprioceptors. Some examples of technical implementation for physical modeling are reviewed. They are software and hardware tools like event-related modeling algorithms, reduced neuron models, robotic drives constructions. We observe existing drives technologies and prospective electric motor types: switched reluctance and transverse flux motors. Next, we look at the existing examples and approaches for robotic modeling of the cerebellum and spinal cord neural networks. These examples show practical methods for the model neural network architecture and adaptation. Those methods allow the use of cortical and spinal cord reflexes for the network training and apply additional artificial blocks for data processing in other brain structures that transmit and receive data from biologically realistic models.

Durable Pneumatic Artificial Muscles with Electric Conductivity for Reliable Physical Reservoir Computing

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.

Robust Multimodal Indirect Sensing for Soft Robots Via Neural Network-Aided Filter-Based Estimation

Sensory data are critical for soft robot perception. However, integrating sensors to soft robots remains challenging due to their inherent softness. An alternative approach is indirect sensing through an estimation scheme, which uses robot dynamics and available measurements to estimate variables that would have been measured by sensors. Nevertheless, developing an adequately effective estimation scheme for soft robots is not straightforward. First, it requires a mathematical model; modeling of soft robots is analytically demanding due to their complex dynamics. Second, it should perform multimodal sensing for both internal and external variables, with minimal sensors, and finally, it must be robust against sensor faults. In this article, we propose a recurrent neural network-based adaptive unscented Kalman filter (RNN-AUKF) architecture to estimate the proprioceptive state and exteroceptive unknown input of a pneumatic-based soft finger. To address the challenge in modeling soft robots, we adopt a data-driven approach using RNNs. Then, we interconnect the AUKF with an unknown input estimator to perform multimodal sensing using a single embedded flex sensor. We also prove mathematically that the estimation error is bounded with respect to sensor degradation (noise and drift). Experimental results show that the RNN-AUKF achieves a better overall performance in terms of accuracy and robustness against the benchmark method. The proposed scheme is also extended to a multifinger soft gripper and is robust against out-of-distribution sensor dynamics. The outcomes of this research have immense potentials in realizing a robust multimodal indirect sensing in soft robots.

Physical reservoir computing with origami and its application to robotic crawling

A new paradigm called physical reservoir computing has recently emerged, where the nonlinear dynamics of high-dimensional and fixed physical systems are harnessed as a computational resource to achieve complex tasks. Via extensive simulations based on a dynamic truss-frame model, this study shows that an origami structure can perform as a dynamic reservoir with sufficient computing power to emulate high-order nonlinear systems, generate stable limit cycles, and modulate outputs according to dynamic inputs. This study also uncovers the linkages between the origami reservoir's physical designs and its computing power, offering a guideline to optimize the computing performance. Comprehensive parametric studies show that selecting optimal feedback crease distribution and fine-tuning the underlying origami folding designs are the most effective approach to improve computing performance. Furthermore, this study shows how origami's physical reservoir computing power can apply to soft robotic control problems by a case study of earthworm-like peristaltic crawling without traditional controllers. These results can pave the way for origami-based robots with embodied mechanical intelligence.

Modelling and implementation of soft bio-mimetic turtle using echo state network and soft pneumatic actuators

Advances of soft robotics enabled better mimicking of biological creatures and closer realization of animals' motion in the robotics field. The biological creature's movement has morphology and flexibility that is problematic deportation to a bio-inspired robot. This paper aims to study the ability to mimic turtle motion using a soft pneumatic actuator (SPA) as a turtle flipper limb. SPA's behavior is simulated using finite element analysis to design turtle flipper at 22 different geometrical configurations, and the simulations are conducted on a large pressure range (0.11-0.4 Mpa). The simulation results are validated using vision feedback with respect to varying the air pillow orientation angle. Consequently, four SPAs with different inclination angles are selected to build a bio-mimetic turtle, which is tested at two different driving configurations. The nonlinear dynamics of soft actuators, which is challenging to model the motion using traditional modeling techniques affect the turtle's motion. Conclusively, according to kinematics behavior, the turtle motion path is modeled using the Echo State Network (ESN) method, one of the reservoir computing techniques. The ESN models the turtle path with respect to the actuators' rotation motion angle with maximum root-mean-square error of [Formula: see text]. The turtle is designed to enhance the robot interaction with living creatures by mimicking their limbs' flexibility and the way of their motion.

Position Control for Soft Actuators, Next Steps toward Inherently Safe Interaction

Soft robots present an avenue toward unprecedented societal acceptance, utility in populated environments, and direct interaction with humans. However, the compliance that makes them attractive also makes soft robots difficult to control. We present two low-cost approaches to control the motion of soft actuators in applications common in human-interaction tasks. First, we present a passive impedance approach, which employs restriction to pneumatic channels to regulate the inflation/deflation rate of a pneumatic actuator and eliminate the overshoot/oscillation seen in many underdamped silicone-based soft actuators. Second, we present a visual servoing feedback control approach. We present an elastomeric pneumatic finger as an example system on which both methods are evaluated and compared to an uncontrolled underdamped actuator. We perturb the actuator and demonstrate its ability to increase distal curvature around the obstacle and maintain the desired end position. In this approach, we use the continuum deformation characteristic of soft actuators as an advantage for control rather than a problem to be minimized. With their low cost and complexity, these techniques present great opportunity for soft robots to improve human–robot interaction.

Skeletonizing the Dynamics of Soft Continuum Body from Video

Soft continuum bodies have demonstrated their effectiveness in generating flexible and adaptive functionalities by capitalizing on the rich deformability of soft material. Compared with a rigid-body robot, it is in general difficult to model and emulate the morphology dynamics of a soft continuum body. In addition, a soft continuum body potentially has an infinite degree of freedom, requiring considerable labor to manually annotate its dynamics from external sensory data such as video. In this study, we propose a novel noninvasive framework for automatically extracting the skeletal dynamics from video of a soft continuum body and show the applications and effectiveness of our framework. First, we demonstrate that our framework can extract skeletal dynamics from animal videos, which can be effectively utilized for the analysis of soft continuum body including animal motion. Next, we focus on a soft continuum arm, a commonly used platform in soft robotics, and evaluate the potential information-processing capability. Normally, to control such a high-dimensional system, it is necessary to introduce many sensors to completely capture the motion dynamics, causing the deterioration of the material's softness. We illustrate that the evaluation of the memory capacity and sensory reconstruction error enables us to verify the minimum number of sensors sufficient for fully grasping the state dynamics, which is highly useful in designing a sensor arrangement for a soft robot. Also, we release the software developed in this study as open source for biology and soft robotics communities, which contributes to automating the annotation process required for the motion analysis of soft continuum bodies.

Review of machine learning methods in soft robotics

Soft robots have been extensively researched due to their flexible, deformable, and adaptive characteristics. However, compared to rigid robots, soft robots have issues in modeling, calibration, and control in that the innate characteristics of the soft materials can cause complex behaviors due to non-linearity and hysteresis. To overcome these limitations, recent studies have applied various approaches based on machine learning. This paper presents existing machine learning techniques in the soft robotic fields and categorizes the implementation of machine learning approaches in different soft robotic applications, which include soft sensors, soft actuators, and applications such as soft wearable robots. An analysis of the trends of different machine learning approaches with respect to different types of soft robot applications is presented; in addition to the current limitations in the research field, followed by a summary of the existing machine learning methods for soft robots.

Sensuator: A Hybrid Sensor–Actuator Approach to Soft Robotic Proprioception Using Recurrent Neural Networks

Soft robotic actuators are now being used in practical applications; however, they are often limited to open-loop control that relies on the inherent compliance of the actuator. Achieving human-like manipulation and grasping with soft robotic actuators requires at least some form of sensing, which often comes at the cost of complex fabrication and purposefully built sensor structures. In this paper, we utilize the actuating fluid itself as a sensing medium to achieve high-fidelity proprioception in a soft actuator. As our sensors are somewhat unstructured, their readings are difficult to interpret using linear models. We therefore present a proof of concept of a method for deriving the pose of the soft actuator using recurrent neural networks. We present the experimental setup and our learned state estimator to show that our method is viable for achieving proprioception and is also robust to common sensor failures.

Dome-Patterned Metamaterial Sheets

The properties of conventional materials result from the arrangement of and the interaction between atoms at the nanoscale. Metamaterials have shifted this paradigm by offering property control through structural design at the mesoscale, thus broadening the design space beyond the limits of traditional materials. A family of mechanical metamaterials consisting of soft sheets featuring a patterned array of reconfigurable bistable domes is reported here. The domes in this metamaterial architecture can be reversibly inverted at the local scale to generate programmable multistable shapes and tunable mechanical responses at the global scale. By 3D printing a robotic gripper with energy-storing skin and a structure that can memorize and compute spatially-distributed mechanical signals, it is shown that these metamaterials are an attractive platform for novel mechanologic concepts and open new design opportunities for structures used in robotics, architecture, and biomedical applications.

Algebraic approach towards the exploitation of “softness”: the input–output equation for morphological computation

Recently, soft robots that consist of soft and deformable materials have received much attention for their adaptability to uncertain environments. Although these robots are difficult to control with a conventional control theory owing to their complex body dynamics, research from different perspectives attempts to actively exploit these body dynamics as an asset rather than a drawback. This approach is called morphological computation, in which the soft materials are used for computation that includes a new kind of control strategy. In this article, we propose a novel approach to analyze the computational properties of soft materials based on an algebraic method, called the input–output equation used in systems analysis, particularly in systems biology. We mainly focus on the two scenarios relevant to soft robotics, that is, analysis of the computational capabilities of soft materials and design of the input force to soft devices to generate the target behaviors. The input–output equation directly describes the relationship between inputs and outputs of a system, and hence by using this equation, important properties, such as the echo state property that guarantees reproducible responses against the same input stream, can be investigated for soft structures. Several application scenarios of our proposed method are demonstrated using typical soft robotic settings in detail, including linear/nonlinear models and hydrogels driven by chemical reactions.

TMTDyn: A Matlab package for modeling and control of hybrid rigid–continuum robots based on discretized lumped systems and reduced-order models

A reliable, accurate, and yet simple dynamic model is important to analyzing, designing, and controlling hybrid rigid–continuum robots. Such models should be fast, as simple as possible, and user-friendly to be widely accepted by the ever-growing robotics research community. In this study, we introduce two new modeling methods for continuum manipulators: a general reduced-order model (ROM) and a discretized model with absolute states and Euler–Bernoulli beam segments (EBA). In addition, a new formulation is presented for a recently introduced discretized model based on Euler–Bernoulli beam segments and relative states (EBR). We implement these models in a Matlab software package, named TMTDyn, to develop a modeling tool for hybrid rigid–continuum systems. The package features a new high-level language (HLL) text-based interface, a CAD-file import module, automatic formation of the system equation of motion (EOM) for different modeling and control tasks, implementing Matlab C-mex functionality for improved performance, and modules for static and linear modal analysis of a hybrid system. The underlying theory and software package are validated for modeling experimental results for (i) dynamics of a continuum appendage, and (ii) general deformation of a fabric sleeve worn by a rigid link pendulum. A comparison shows higher simulation accuracy (8–14% normalized error) and numerical robustness of the ROM model for a system with a small number of states, and computational efficiency of the EBA model with near real-time performances that makes it suitable for large systems. The challenges and necessary modules to further automate the design and analysis of hybrid systems with a large number of states are briefly discussed.

Magnetic Shape Memory Polymers with Integrated Multifunctional Shape Manipulation

Shape-programmable soft materials that exhibit integrated multifunctional shape manipulations, including reprogrammable, untethered, fast, and reversible shape transformation and locking, are highly desirable for a plethora of applications, including soft robotics, morphing structures, and biomedical devices. Despite recent progress, it remains challenging to achieve multiple shape manipulations in one material system. Here, a novel magnetic shape memory polymer composite is reported to achieve this. The composite consists of two types of magnetic particles in an amorphous shape memory polymer matrix. The matrix softens via magnetic inductive heating of low-coercivity particles, and high-remanence particles with reprogrammable magnetization profiles drive the rapid and reversible shape change under actuation magnetic fields. Once cooled, the actuated shape can be locked. Additionally, varying the particle loadings for heating enables sequential actuation. The integrated multifunctional shape manipulations are further exploited for applications including soft magnetic grippers with large grabbing force, reconfigurable antennas, and sequential logic for computing.

Optimal short-term memory before the edge of chaos in driven random recurrent networks

The ability of discrete-time nonlinear recurrent neural networks to store time-varying small input signals is investigated with mean-field theory. The combination of a small input strength and mean-field assumptions makes it possible to derive an approximate expression for the conditional probability density of the state of a neuron given a past input signal. From this conditional probability density, we can analytically calculate short-term memory measures, such as memory capacity, mutual information, and Fisher information, and determine the relationships among these measures, which have not been clarified to date to the best of our knowledge. We show that the network contribution of these short-term memory measures peaks before the edge of chaos, where the dynamics of input-driven networks is stable but corresponding systems without input signals are unstable.

The design and development of Branch Bot: a branch-crawling, caterpillar-inspired, soft robot

A soft climbing robot has the potential to access locations such as wiring ducts and tree canopies that are unreachable by humans and traditional rigid robots. In addition, a soft robot is robust and can fall without damaging itself or its environment. We present a soft, branch-crawling robot that is inspired by the passive gripping mechanisms used by caterpillars. The conformability of the robot’s soft body makes it uniquely suited to move in a complex 3D environment. A key innovation is that grip release is actively controlled and coordinated with propulsion generated by stored elastic energy. The robot is molded from silicone rubber and actuated using remote motor-tendons coupled to the structure through Bowden cables. Grip is achieved passively through an elastic flexure that pushes a compliant finger against the dowel. Experimental results show that the gripper is easily able to support the weight of the robot, and that the body structure allows the robot to crawl horizontally, vertically, and along branches. This robot demonstrates some key advantages of a soft robotic platform over traditional rigid robots.