Ryan L. Truby

Abstract This perspective is based on a talk titled, “Soft Architected Robots: Printing Complex Forms for New Sensorimotor Capabilities in Robotics,” presented at the Fall 2024 Meeting of the Materials Research Society as part of the “Distinguished Invited Speaker” series. We highlight the latest progress in developing architected materials—namely kirigami, origami, auxetic structures, and lattices—for soft robots. In particular, we focus on recent examples of using soft, architected materials for soft robotic actuators, sensors, and sensorized soft actuators with embedded sensing capabilities. We provide an outlook on emerging opportunities in the use, design, and manufacturing of architected materials to advance the capabilities and practical performance of soft robots. We encourage the field to see this class of materials as essential to advancing robot capabilities more broadly beyond those afforded by traditional means and mechanisms. Graphical abstract

Soft robotic actuators present tradeoffs between mechanical compliance and practical capabilities, motivating new actuation strategies that harness the storage and rapid release of elastic energy for high-performance movements. We present a servo-driven, soft linear actuator that integrates compact mechanical clutches with architected elastomers based on handed shearing auxetics (HSAs). The actuator stores elastic strain energy during servo-driven HSA extension that can be released upon clutch disengagement to drive rapid contraction. Altogether, the actuator achieves a max pulling force of 184 N, actuation strain of 23%, and contraction speed of 1.55 m s-1. A comparative characterization of performance with and without clutching reveals a nearly ten times increase in peak power with clutch-enabled actuation, while maintaining intrinsic compliance and back-drivability. When integrated into an artificial musculoskeletal system like a human-scale robot leg, the actuators enable dynamic tasks through clutch disengagement, such as ball kicking and ground push-off. Rapid clutch re-engagement can also stabilize motion by suppressing rebound oscillations. We evaluate the leg’s performance in these tasks, including transferred mechanical work, energy efficiency, and force ouput. The results demonstrate the actuator’s ability to generate fast, powerful motions and potential to advance artificial musculoskeletal systems for bioinspired robots.

The development of advanced materials capable of performing multiple functions is a key step toward adaptive, autonomous systems for emerging technologies. However, multifunctional material systems designed to integrate sensory, computational, and actuation capabilities are challenging to realize due to manufacturing and materials limitations. Here, we present electrically controllable, multifunctional iontronic composites (MICs) that demonstrate ionic sensing, current regulation, and ionomotive bending actuation capabilities within a single architecture. Our MICs are fabricated using a multimaterial direct-write printing process, in which a poly(ionic liquid) (pIL) structural electrolyte is sandwiched between two Ti 3 C 2 T x MXene-based electrodes. The printing process enables seamless integration of concentrated MXene electrode inks with pIL electrolytes with printable layer thicknesses down to 25 and 200 μm, respectively. When used as a sensor, MICs exhibit capacitance changes up to 4% under compressive loads of 45 N. When printed with electrodes of asymmetric thickness, MICs can also function as ionic diodes, achieving rectification ratios up to 14. Finally, the composites demonstrate ionomotive actuation with a maximum bending strain of 0.21%. Our key innovation lies in achieving all three functionalities through additive manufacturing, which reduces the number of fabrication steps required to integrate all MIC materials together. Our MICs represent a significant advance in electrically controlled, multifunctional composites and motivate new directions toward next-generation autonomous and responsive material systems for soft robotics, electronics, and adaptive structures.

Closed-loop control remains an open challenge in soft robotics. The nonlinear responses of soft actuators under dynamic loading conditions limit the use of analytic models for soft robot control. Traditional methods of controlling soft robots underutilize their configuration spaces to avoid nonlinearity, hysteresis, large deformations, and the risk of actuator damage. Furthermore, episodic data-driven control approaches such as reinforcement learning (RL) are traditionally limited by sample efficiency and inconsistency across initializations. In this work, we demonstrate RL for reliably learning control policies for dynamic balancing tasks in real-time single-shot hardware deployments. We use a deformable Stewart platform constructed using parallel, 3D-printed soft actuators based on motorized handed shearing auxetic (HSA) structures. By introducing a curriculum learning approach based on expanding neighborhoods of a known equilibrium, we achieve reliable single-deployment balancing at arbitrary coordinates. In addition to benchmarking the performance of model-based and model-free methods, we demonstrate that in a single deployment, Maximum Diffusion RL is capable of learning dynamic balancing after half of the actuators are effectively disabled, by inducing buckling and by breaking actuators with bolt cutters. Training occurs with no prior data, in as fast as 15 minutes, with performance nearly identical to the fully-intact platform. Single-shot learning on hardware facilitates soft robotic systems reliably learning in the real world and will enable more diverse and capable soft robots.

Soft Actuators In their Research Article (DOI: 10.1002/adma.202501290), Ryan L. Truby and co-workers construct architected soft actuators from a combination of 3D printed elastomers that extend and contract upon rotation of integrated servo motors. The actuators exhibit high actuation stroke, force output, and power density. The authors construct human-scale legs with bone-inspired links, tendons, and three architected soft actuators. Image credit: T. Kim, R. L. Truby, Northwestern University.

Closed-loop control remains an open challenge in soft robotics. The nonlinear responses of soft actuators under dynamic loading conditions limit the use of analytic models for soft robot control. Traditional methods of controlling soft robots underutilize their configuration spaces to avoid nonlinearity, hysteresis, large deformations, and the risk of actuator damage. Furthermore, episodic data-driven control approaches such as reinforcement learning (RL) are traditionally limited by sample efficiency and inconsistency across initializations. In this work, we demonstrate RL for reliably learning control policies for dynamic balancing tasks in real-time single-shot hardware deployments. We use a deformable Stewart platform constructed using parallel, 3D-printed soft actuators based on motorized handed shearing auxetic (HSA) structures. By introducing a curriculum learning approach based on expanding neighborhoods of a known equilibrium, we achieve reliable single-deployment balancing at arbitrary coordinates. In addition to benchmarking the performance of model-based and model-free methods, we demonstrate that in a single deployment, Maximum Diffusion RL is capable of learning dynamic balancing after half of the actuators are effectively disabled, by inducing buckling and by breaking actuators with bolt cutters. Training occurs with no prior data, in as fast as 15 minutes, with performance nearly identical to the fully-intact platform. Single-shot learning on hardware facilitates soft robotic systems reliably learning in the real world and will enable more diverse and capable soft robots.

Responsive materials offer solutions to complex engineering challenges by enabling systems to adapt their shapes or properties in response to external stimuli. To fully harness the potential of responsive materials, inverse design methods that integrate multiple types of stimuli and manufacturing processes are necessary. We present a unified, autonomous codesign framework that simultaneously optimizes structure, manufacturing, materials, and stimuli for responsive material systems, achieving target shape morphing under multiple stimuli without relying on human heuristics or expertise. It integrates generalized topology optimization with hybrid data-physics differentiable simulations to achieve flexible, manufacturing-aware designs for network-like responsive material systems. We showcase our framework with a multimaterial three-dimensional printing process with high material tunability, which we use to fabricate liquid crystal elastomer systems that morph into different forms in response to heat and light. The exceptional flexibility and efficiency of our method will advance shape-morphing applications spanning soft robotics to drug delivery.

Vertebrates depend on their musculoskeletal system for locomotion, manipulation, interaction with their environment, and more. The robustness and efficiency of animal locomotion are difficult to achieve in robots because their hardware does not replicate the mechanics and performance of animal bodies. Moreover, many state-of-the-art soft actuators are ill-suited as muscles in artificial musculoskeletal systems for deployable, task-capable robots. This study presents an electrically-driven, architected soft actuator that can be assembled into artificial musculoskeletal systems. The fully 3D printed actuators linearly extend and contract through the rotation of an integrated servo motor. They comprise a thermoplastic polyurethane handed shearing auxetic (HSA) and origami bellows structure. Together, these structures transmit torque, stretch, and resist torsional deflection in a manner that produces large linear actuation and force output up to 59 mm (or 30% strain) and 75 N, respectively. It showcases the actuator's performance as artificial muscles in a battery-powered, human-scale leg that can use three muscles to kick a ball. When accounting for the weight of auxiliary hardware, the actuators exhibit power and energy densities that are four orders of magnitude higher than for leading soft artificial muscles. The soft actuators represent a step toward providing robots with bioinspired musculoskeletal systems for animal-like abilities.

Abstract Developing fully untethered soft robots that leverage mechanically compliant bodies for robust, adaptable, and bioinspired performance remains a grand challenge. State‐of‐the‐art soft actuators pose major limitations in their force output, power density, energy efficiency, and dependence on bulky, heavy power supplies. These challenges are addressed with electrically driven soft actuators based on motorized, 3D printed architected elastomers. The presented actuators are fully 3D printed from thermoplastic polyurethane (TPU) and have a handed shearing auxetic (HSA) structure. The soft robotic legs are flexible, durable, and driven directly by integrated servo motors. The structure‐property‐performance relationships of HSA legs are explored for soft robot walking by varying the auxetic pattern region of HSAs. Through characterizations of HSA leg performance and soft robot walking, it is demonstrated that soft robotic quadrupeds with the most mechanically compliant legs achieve the fastest walking speed (183 mm s −1 or 0.65 body lengths per second) and the lowest cost of transport—despite generating the lowest forces. It is anticipated that this work will spark new directions in the optimized design and manufacturing of soft actuators for untethered locomotion and the creation of deployable soft robots that can practically operate in unstructured, real‐world environments.

Many soft robots struggle to produce dynamic motions with fast, large displacements. We develop a parallel 6 degree-of-freedom (DoF) Stewart-Gough mechanism using Handed Shearing Auxetic (HSA) actuators. By using soft actuators, we are able to use one third as many mechatronic components as a rigid Stewart platform, while retaining a working payload of 2kg and an open-loop bandwidth greater than 16Hz. We show that the platform is capable of both precise tracing and dynamic disturbance rejection when controlling a ball and sliding puck using a Proportional Integral Derivative (PID) controller. We develop a machine-learning-based kinematics model and demonstrate a functional workspace of roughly 10cm in each translation direction and 28 degrees in each orientation. This 6DoF device has many of the characteristics associated with rigid components—power, speed, and total workspace— while capturing the advantages of soft mechanisms.

Handed shearing auxetics, or HSAs, are a class of architected materials increasingly used as electrically-driven soft actuators. HSAs are directly driven by servo motors, resulting in architected soft robotic actuators that enable capabilities spanning manipulation and locomotion. However, the material properties and form factors available to HSAs are limited. Thus, fabricating miniaturized HSAs from robust, durable materials is difficult. Moreover, scaling HSAs to smaller form factors is also complicated by the need to miniaturize the motors driving them. Here, we present a method for fabricating miniaturized, robust HSA actuators via molding from thermoplastic polyurethane (TPU) powders. Our method produces soft HSA actuators with low torque requirements that can be actuated with DC micromotors. We describe the overall fabrication process for our actuators, characterize the free displacement and blocked force generated by single HSAs, and demonstrate the performance of a multi-DoF platform comprising a 2x2 assembly of HSAs. We find that our new HSAs produce actuation strains and forces of 40% and 1.2N, respectively, with servo motors; with DC micromotors, they can actuate to at least 20Hz. Altogether, the use of micromotors and thermoplastic elastomers enables us to achieve extremely robust and fast actuation with HSAs. We expect our new approach to HSA design, fabrication, and actuation will open up new opportunities in the use of architected soft robotic actuators that operate with the actuation bandwidths found in both rigid robots and living organisms.

Tails are flexible appendages that many vertebrates use for balance, gait stabilization, thrust generation, and more. While robots rarely have them, soft walking robots may benefit from a tail that stabilizes locomotion in unstructured terrains. We present a tendon-driven, soft robotic tail capable of swinging and shortening to change the momentum and center of mass of a robot body. The tail comprises three origami bellows fully 3D printed from thermoplastic polyurethane. The bellows are highly compressible, allowing motorized tendon actuators to shorten the tail by 110 mm, or 30% of the tail’s initial length. The tail’s flexibility also enables large swinging motions, which are achieved by alternatively pulling and releasing two tendons routing at the sides of the bellows structures. Since the tail’s servo motors are at its end, the tail can generate up to 0.5 N•m torque while swinging. In this paper, we characterize the tail’s range of adjustable lengths, the swinging motions it can produce, and the torque it generates. Swinging and torque generation are evaluated at several different initial tail lengths. Finally, we demonstrate the tail’s controllability through a closed proportional-integral-derivative (PID) feedback controller. This work sets in motion future investigations of how vertebrate-inspired tails can enhance the mobility and stability of (soft) robot walking.

A Flexible, Architected Soft Robotic Actuator for Motorized Extensional Motion In article number 2300866 by Ryan L. Truby and co-workers, a flexible, architected soft robotic actuator is reported that comprises a 3D printed, cylindrical handed shearing auxetic structure and a deformable, internal rubber bellows shaft. The actuator linearly extends upon applying torque from a servo motor, while maintaining high flexibility. The photomontage shows the soft actuator used as a crawling soft robot that can move through tight, tortuous surroundings. The mechanical deformability allows the actuator to passively adapt to its environment.

To advance the design space of electrically‐driven soft actuators, a flexible, architected soft robotic actuator is presented for motor‐driven extensional motion. The actuator comprises a 3D printed, cylindrical handed shearing auxetic (HSA) structure and a deformable, internal rubber bellows shaft. The actuator linearly extends upon applying torque from a servo motor; the rubber bellows shaft is stretchable but resistant to torsional deflection, allowing it to transmit torque from the servo motor to the other end of the HSA. The high flexibility of the HSA and rubber bellows shaft enable the actuator to adaptively extend even when bent. The actuator's two components and its performance are mechanically characterized. Actuation strains of 45% elongation and a maximum blocked pushing force of about 8 N are demonstrated. The actuator's capabilities are showcased in two separate demonstrations: a crawling robot and a sensorized artificial muscle that integrates a microfluidic, liquid metal strain sensor. The architected material design approach for a robust, motor‐driven soft actuator provides several unique features—including a compact form factor and ease of use—over other motorized soft robotic actuators based on HSA assemblies or cable tendon mechanisms.

Structural electrolytes present advantages over liquid varieties, which are critical to myriad applications. In particular, structural electrolytes based on polymerized ionic liquids or poly(ionic liquids) (pILs) provide wide electrochemical windows, high thermal stability, nonvolatility, and modular chemistry. However, current methods of fabricating structural electrolytes from pILs and their composites present limitations. Recent advances have been made in 3D printing pIL electrolytes, but current printing techniques limit the complexity of forms that can be achieved, as well as the ability to control mechanical properties or conductivity. We introduce a method for fabricating architected pIL composites as structural electrolytes via embedded 3D (EMB3D) printing. We present a modular design for formulating ionic liquid (IL) monomer composite inks that can be printed into sparse, lightweight, free-standing lattices with different functionalities. In addition to characterizing the rheological and mechanical behaviors of IL monomer inks and pIL lattices, we demonstrate the self-sensing capabilities of our printed structural electrolytes during cyclic compression. Finally, we use our inks and printing method to spatially program self-sensing capabilities in pIL lattices through heterogeneous architectures as well as ink compositions that provide mixed ionic-electronic conductivity. Our free-form approach to fabricating structural electrolytes in complex, 3D forms with programmable, anisotropic properties has broad potential use in next-generation sensors, soft robotics, bioelectronics, energy storage devices, and more.

Machine learning-based approaches for soft robot proprioception have recently gained popularity, in part due to the difficulties in modeling the relationship between sensor signals and robot shape. However, to date, there exists no systematic analysis of the required design choices to set up a machine learning pipeline for soft robot proprioception. Here, we present the first study examining how design choices on different levels of the machine learning pipeline affect the performance of a neural network for predicting the state of a soft robot. We address the most frequent questions researchers face, such as how to choose the appropriate sensor and actuator signals, process input and output data, deal with time series, and pick the best neural network architecture. By testing our hypotheses on data collected from two vastly different systems–an electrically actuated robotic platform and a pneumatically actuated soft trunk–we seek conclusions that may generalize beyond one specific type of soft robot and hope to provide insights for researchers to use machine learning for soft robot proprioception.

Gait generation for soft robots is challenging due to the nonlinear dynamics and high dimensional input spaces of soft actuators. Limitations in soft robotic control and perception force researchers to hand-craft open loop controllers for gait sequences, which is a non-trivial process. Moreover, short soft actuator lifespans and natural variations in actuator behavior limit machine learning techniques to settings that can be learned on the same time scales as robot deployment. Lastly, simulation is not always possible, due to heterogeneity and nonlinearity in soft robotic materials and their dynamics change due to wear. We present a sample-efficient, simulation free, method for self-generating soft robot gaits, using very minimal computation. This technique is demonstrated on a motorized soft robotic quadruped that walks using four legs constructed from 16 “handed shearing auxetic” (HSA) actuators. To manage the dimension of the search space, gaits are composed of two sequential sets of leg motions selected from 7 possible primitives. Pairs of primitives are executed on one leg at a time; we then select the best-performing pair to execute while moving on to subsequent legs. This method-which uses no simulation, sophisticated computation, or user input-consistently generates good translation and rotation gaits in as low as 4 minutes of hardware experimentation, outperforming hand-crafted gaits. This is the first demonstration of completely autonomous gait generation in a soft robot.

Gait generation for soft robots is challenging due to the nonlinear dynamics and high dimensional input spaces of soft actuators. Limitations in soft robotic control and perception force researchers to hand-craft open loop controllers for gait sequences, which is a non-trivial process. Moreover, short soft actuator lifespans and natural variations in actuator behavior limit machine learning techniques to settings that can be learned on the same time scales as robot deployment. Lastly, simulation is not always possible, due to heterogeneity and nonlinearity in soft robotic materials and their dynamics change due to wear. We present a sample-efficient, simulation free, method for self-generating soft robot gaits, using very minimal computation. This technique is demonstrated on a motorized soft robotic quadruped that walks using four legs constructed from 16 "handed shearing auxetic" (HSA) actuators. To manage the dimension of the search space, gaits are composed of two sequential sets of leg motions selected from 7 possible primitives. Pairs of primitives are executed on one leg at a time; we then select the best-performing pair to execute while moving on to subsequent legs. This method -- which uses no simulation, sophisticated computation, or user input -- consistently generates good translation and rotation gaits in as low as 4 minutes of hardware experimentation, outperforming hand-crafted gaits. This is the first demonstration of completely autonomous gait generation in a soft robot.

A combustion-powered soft actuator takes microrobots to new heights and speeds.

The impressive capabilities of living organisms arise from the way autonomy is materialized by their bodies. Across scales, living beings couple computational or cognitive intelligence with physical intelligence through body morphology, material multifunctionality, and mechanical compliance. While soft robotics has advanced the design and fabrication of physically intelligent bodies, the integration of information‐processing capabilities for computational intelligence remains a challenge. Consequently, perception and control limitations have constrained how soft robots are built today. Progress toward untethered autonomy will require deliberate convergence in how the field codevelops new materials, fabrication methods, and control strategies for soft robots. Here, a new perspective is put forward: that researchers should use tasks alone to impose material and information constraints on soft robot design. A conceptual framework is proposed for a task‐first design paradigm that sidesteps limitations imposed by control strategies. This framework allows emergent synergies between material and information processing properties of soft matter to be readily exploited for task‐capable agents. Particular attention is paid to the scale dependence of solutions. Finally, an outlook is presented on emerging research opportunities for achieving autonomy in future soft robots as large as elephant trunks and as small as paramecia.

Here we highlight 35 (+1) global researchers approximately under the age of 35. The annual cohort was self-generated by initial seed invitations sent by the editorial team, with each contributor inviting the next in a self-selecting unrestricted (nominally supervised) manner. The final collection is an inspiring look at the challenges the current generation of materials researchers are tackling, demonstrating the interdisciplinarity of materials science.

Adv. Mater. 2023, 35, 2206958 DOI: 10.1002/adma.202206958 In the acknowledged funding for the published article, the grant numbers for NSF through Harvard Materials Research Science and Engineering Center Grant (MRSEC) and NSF Designing Materials to Revolutionize and Engineer our Future were incorrect. These grant numbers are hereby corrected below: NSF through Harvard Materials Research Science and Engineering Center Grant (MRSEC): Grant no. DMR-2011754 NSF Designing Materials to Revolutionize and Engineer our Future: DMREF-1922321 The acknowledgements text is corrected as: R.D.W. and R.L.T. contributed equally to this work. The authors acknowledge support from the NSF through Harvard Materials Research Science and Engineering Center Grant (MRSEC) DMR-2011754 and NSF Designing Materials to Revolutionize and Engineer our Future (DMREF-1922321). J.A.L. also thanks GETTYLAB for their generous support of this work. Finally, the authors thank L. K. Sanders for assistance with photography and videography.

Soft robotics aims to close the performance gap between built and biological machines through materials design. Soft robots are constructed from soft, actuatable materials to be physically intelligent, or to have traits that living organisms possess such as passive adaptability and morphological computation through their compliant, deformable bodies. However, materials selection for physical intelligence often involves low-performance and/or energy-inefficient, stimuli-responsive materials for actuation. Additional challenges in soft robot sensorization and control further limit the practical utility of these machines. Recognizing that electrically controllable materials are crucial for the development of soft machines that are both physically and computationally intelligent, we review progress in the development of electroprogrammable materials for soft robotic actuation. We focus on thermomechanical, electrostatic, and electrochemical actuation strategies that are directly controlled by electric currents and fields. We conclude with an outlook on the design and fabrication of next-generation robotic materials that will facilitate true bioinspired autonomy.

Electrically-actuated continuum soft robots based on Handed Shearing Auxetics (HSAs) promise rapid actuation capabilities while preserving structural compliance. However, the foundational models of these novel actuators required for precise control strategies are missing. This paper proposes two key components extending discrete Cosserat rod model (DCM) to allow for modeling HSAs. First, we propose a mechanism for incorporating the auxetic trajectory into DCM dynamical simulations. We also propose an implementation of this extension as a plugin for the Elastica simulator. Second, we introduce a Selective Piecewise Constant Strain (SPCS) kinematic parameterization that can describe an HSA segment's shape with fewer configuration variables. We verify both theoretical contributions experimentally. The simulator is used to replicate experimental data of the mechanical characterization of HSA rods. For the second component, we attach motion capture markers at various points to a parallel HSA robot and find that the shape of the HSAs can be kinematically represented with an average accuracy of 0.3 mm for positions and 0.07 rad for orientations.

Light- and ink-based 3D printing methods have vastly expanded the design space and geometric complexity of architected ceramics. However, light-based methods are typically confined to a relatively narrow range of preceramic and particle-laden resins, while ink-based methods are limited in geometric complexity due to layerwise assembly. Here, embedded 3D printing is combined with microwave-activated curing to generate architected ceramics with spatially controlled composition in freeform shapes. Aqueous colloidal inks are printed within a support matrix, rapidly cured via microwave-activated polymerization, and subsequently dried and sintered into dense architectures composed of one or more oxide materials. This integrated manufacturing method opens new avenues for the design and fabrication of complex ceramic architectures with programmed composition, density, and form for myriad applications.