Soft robotics最新文献

Dielectric elastomer actuators (DEAs) enable to create soft robots with fast response speed and high-energy density, but the fast optimization design of DEAs still remains elusive because of their continuous electromechanical deformation and high-dimensional design space. Existing approaches usually involve repeating and vast finite element calculation during the optimization process, leading to low efficiency and time consuming. The advance of deep learning has shown the potential to accelerate the optimization process, but the high-dimensional design space leads to challenge on the accuracy and generality of the deep learning model. In this work, we propose a deep learning-based automatic design framework for DEAs, capable of rapidly generating high-dimensional distributed electrode patterns based on different design objects. This framework is developed as follows: (1) a dataset construction strategy combining with a finite element model is developed to optimize the data distribution within the high-dimensional design space; (2) a neural network-embedded physical information is designed and trained to achieve accurate prediction of the continuous deformation within $0.011s$; and (3) a genetic algorithm with the neural network is proposed to automatically and rapidly optimize the electrode pattern of DEAs based on various design objects. To verify the effectiveness, a series of case studies (including maximum displacement, specific displacement, multiplicity of solutions, multiple degree-of-freedom actuations, and complex actuations) has been conducted. Both simulation results and experimental data demonstrate that our design framework can automatically design the electrode pattern within 2 min and obviously improve the performance of DEAs. This work proposes a deep learning-based design approach with automatic and rapid property, thereby paving the way for broader applications of DEAs.

Tendon-driven continuum robots suffer from crosstalk of driving forces between sections, typically resulting in motion coupling between sections, which affects their motion accuracy and complicates the control strategies. To address these issues, this article proposes a mechanically designed variable-pitch flexible-screw-driven continuum robot (FSDCR) that enables motion decoupling between sections. The continuum section of the FSDCR comprises a series of orthogonally arranged vertebrae and is driven by customized variable-pitch flexible screws. The variable-pitch flexible screws apply driving forces and constraints to several threaded vertebrae in the continuum section, improving positioning accuracy and loading capacity. The flexible screws effectively balance the driving force and torque within one section through antagonistic torsional actuation, thereby achieving motion decoupling between sections. Characterization experiments have been conducted to compare the motion accuracy and load capacity of the variable-pitch FSDCR with those of the constant-pitch FSDCR. The results demonstrate that the variable-pitch FSDCR exhibits improved positioning accuracy, minimizing an average error of 0.79 mm (0.60% relative to its total length), which is 82.09% lower than that of the constant-pitch FSDCR. The load capacity of the variable-pitch FSDCR is enhanced by up to 129.09% compared with the constant-pitch FSDCR. Experiments on the motion decoupling performance of the FSDCR show that the maximum motion coupling error is 0.32 mm (0.24% relative to the section length). Additionally, the motion coupling error is minimally influenced by the rotational speed of the screw. Finally, a three-section FSDCR is constructed, and its load capacity and motion flexibility are demonstrated.

High-performance eco-friendly soft actuators showing large displacement, fast response, and long-term operational capability require further development for next-generation bioinspired soft robots. Herein, we report an electro-ionic soft actuator based on carboxylated cellulose nanocrystals (CCNC) and carboxylated cellulose nanofibers (CCNF), graphene nanoplatelets (GN), and ionic liquid (IL). The actuator exhibited exceptional actuation performances, achieving large displacements ranging from 1.6 to 12.3 mm under ultralow actuation voltages of 0.25-1.5 V. It also operated stably across a broad frequency band from 0.1 to 10 Hz and displayed a significant working stability of 99.3% after up to 240 cycles. Remarkably, the electro-active actuator demonstrated a fast response (0.39 s delay under 1.0 V at 0.1 Hz), and a long lifespan (with only a minor decrease of 2% for 2 years). The enhanced actuation performances of the actuator were attributed to its superior ionic conductivity, high charge storage ability, strong ionic interaction, and physical-chemical cross-linked networks. Furthermore, we successfully demonstrated the bioinspired applications of CCNC/CCNF-IL-GN actuators including micro-grippers, spiral-structure electroactive stents, biomimetic fingers, and bionic dragonfly wings. The proposed actuator and its bioinspired robot designs could offer a significant way for the development of next-generation eco-friendly soft actuators, soft robots, and biomedical microdevices in microenvironments requiring low-voltage environment.

Soft robotics is gaining interest in rehabilitation applications, bringing new opportunities to offset the loss of upper limb motor function following neurological, neuromuscular, or traumatic injuries. Unlike conventional rigid robotics, the added softness in linkages or joints promises to make rehabilitation robots compliant, which translates into higher levels of safety, comfort, usability, and portability, opening the door for these rehabilitation technologies to be used in daily life. While several reviews documented the different technical implementations of soft rehabilitation robots, it is essential to discuss the growing clinical evidence on the feasibility and effectiveness of using this technology for rehabilitative and assistive purposes, whether softness brings the expected advantages from the perspective of end users, and how we should proceed in the future of this field. In this perspective article, we present recent clinical evidence on how 13 different upper limb devices were used in both controlled (clinical) and uncontrolled (at home) settings in more than 37 clinical studies. From these findings and our own experience, we derive recommendations for future developers and end users regarding the design, application, and evaluation of soft robotics for upper limb rehabilitation and assistance.

Programmable deformation hydrogel robots have garnered significant attention in biomedical fields due to their ability to undergo large-scale reversible deformation. As clinical demand rises, there is a need for hydrogel robots that are easy to process and operate, and can undergo programmable deformation. Here, we propose a method to fabricate single-layer programmable deformation hydrogel robots in one step using a high-precision digital light processing 3D printing system. Two kinds of deformable elements with different structure distribution on the top and bottom sides are produced by using two kinds of focused light with varying intensities. By combining these deformable elements, we create four basic modules with different and fixed deformable shapes. The desired shape deformation in hydrogel robots can be achieved by programming the combination of these four basic modules. The hydrogel robots exhibit reversible repeat deformation under near-infrared light stimulation. We validate our approach by fabricating several scaffolds using combinations of the four basic modules, demonstrating the feasibility of programming deformation and the potential application of these scaffolds in pipeline movement. This research provides the feasibility for the simple programming deformation of hydrogel robots and offers a novel approach for fabricating programmable deformation hydrogel robots in biomedical fields.

Data-driven calibration methods have shown promising results for accurate proprioception in soft robotics. This process can be greatly benefited by adopting numerical simulation for computational efficiency. However, the gap between the simulated and real domains limits the accurate, generalized application of the approach. Herein, we propose an unsupervised domain adaptation framework as a data-efficient, generalized alignment of these heterogeneous sensor domains. A dual cross-modal autoencoder was designed to match the sensor domains at a feature level without any extensive labeling process, facilitating the computationally efficient transferability to various tasks. Moreover, our framework integrates domain adaptation with anomaly detection, which endows robots with the capability for external collision detection. As a proof-of-concept, the methodology was adopted for the famous soft robot design, a multigait soft robot, and two fundamental perception tasks for autonomous robot operation, involving high-fidelity shape estimation and collision detection. The resulting perception demonstrates the digital-twinned calibration process in both the simulated and real domains. The proposed design outperforms the existing prevalent benchmarks for both perception tasks. This unsupervised framework envisions a new approach to imparting embodied intelligence to soft robotic systems via blending simulation.

Wearable robots have been developed to assist the physical performance of humans. Specifically, exosuits have attracted attention due to their lightweight and soft nature, which facilitate user movement. Although several types of force controllers have been used in exosuits, it is challenging to control the assistive force due to the material's softness. In this study, we propose three methods to improve the performance of the basic controller using an admittance-based force controller. In method A, the cable was controlled according to the user's thigh motion to eliminate delays in generating the assistive force and improve the control accuracy. In method B, the stiffness feedforward model of the human exosuit was divided into two independent models based on the assistance phase for compensating the nonlinear stiffness more accurately. In method C, the real-time optimization method for the stiffness feedforward model with an adaptive moment estimation method optimizer was proposed. To validate these methods' effectiveness, we designed three new controllers, gradually combined the proposed methods with the basic controller, and compared their performances. We found that controller III, combining all three methods with the basic controller, showed the best performance. By applying controller III in the same exosuit, the root-mean-square error of the assistive force decreased from 39.84 N to 13.72 N, reducing the error by 65.56% compared with the basic controller. Moreover, the time delay for force generation in the gait cycle percentage decreased from 9.99% to 3.41%, reducing the delay by 65.87% compared with the basic controller.

This paper explores online stiffness modulation within a single tail stroke for swimming soft robots. Despite advances in stiffening mechanisms, little attention has been given to dynamically adjusting stiffness in real-time, presenting a challenge in developing mechanisms with the requisite bandwidth to match tail actuation. Achieving an optimal balance between thrust and efficiency in swimming soft robots remains elusive, and the paper addresses this challenge by proposing a novel mechanism for independent stiffness control, leveraging fluid-driven stiffening within a patterned pouch. Inspired by fluidic-driven actuation, this approach exhibits high bandwidth and facilitates significant stiffness changes. We perform experiments to demonstrate how this mechanism enhances both thrust and swimming efficiency. The tail actuation and fluid-driven stiffening can be optimized for a specific combination of thrust and efficiency, tailored to the desired maneuver type. The paper further explores the complex interaction between the soft body and surrounding fluid and provides fluid dynamics insights gained from the vortices created during actuation. Through frequency modulation and online stiffening, the study extends the Pareto front of achievable thrust generation and swimming efficiency.

Numerous soft actuators based on pneumatic network (PneuNet) design have already been proposed and extensively employed across various soft robotics applications in recent years. Despite their widespread use, a common limitation of most existing designs is that their action is predetermined during the fabrication process, thereby restricting the ability to modify or alter their function during operation. To address this shortcoming, in this article the design of a Reconfigurable, Transformable Soft Pneumatic Actuator (RT-SPA) is proposed. The working principle of the RT-SPA is analogous to the conventional PneuNet. The key distinction between the two lies in the ability of the RT-SPA to undergo controlled transformations, allowing for more versatile bending and twisting motions in various directions. Furthermore, the unique reconfigurable design of the RT-SPA enables the selection of actuation units with different sizes to achieve a diverse range of three-dimensional deformations. This versatility enhances the RT-SPA's potential for adaptation to a multitude of tasks and environments, setting it apart from traditional PneuNet. The article begins with a detailed description of the design and fabrication of the RT-SPA. Following this, a series of experiments are conducted to evaluate the performance of the RT-SPA. Finally, the abilities of the RT-SPA for locomotion, gripping, and object manipulation are demonstrated to illustrate the versatility of the RT-SPA across different aspects.

This article explores the concept of external magnetic control for vine robots to enable their high curvature steering and navigation for use in endoluminal applications. Vine robots, inspired by natural growth and locomotion strategies, present unique shape adaptation capabilities that allow passive deformation around obstacles. However, without additional steering mechanisms, they lack the ability to actively select the desired direction of growth. The principles of magnetically steered growing robots are discussed, and experimental results showcase the effectiveness of the proposed magnetic actuation approach. We present a 25-mm-diameter vine robot with an integrated magnetic tip capsule, including 6 degrees of freedom (DOF) localization system and camera, and demonstrate a minimum bending radius of 3.85 cm with an internal pressure of 30 kPa. Furthermore, we evaluate the robot's ability to form tight curvature through complex navigation tasks, with magnetic actuation allowing for extended free-space navigation without buckling. The suspension of the magnetic tip was also validated using the 6 DOF localization system to ensure that the shear-free nature of vine robots was preserved. Additionally, by exploiting the magnetic wrench at the tip, we showcase preliminary results of vine retraction. The findings contribute to the development of controllable vine robots for endoluminal applications, providing high tip force and shear-free navigation.