Publications

ABSTRACT Designing soft electronic skins for tactile sensing facilitates natural human–machine communications. However, the nonlinear characteristics of electrical transductions and mechanics usually compromise precise tactile decoding and compliant interactions. In this work, we demonstrate a soft 3D‐architectured pressure sensor featuring a PEDOT:PSS‐PVA hydrogel lattice encapsulated within an origami‐inspired elastomeric framework. Our 3D lattice sensor leverages the ultracapacitive principle to achieve wide‐range linearity (0–220 kPa), fast responses, and high‐resolution detection under extreme loading. The proposed 3D configuration also enables linear compression behaviors within ∼49.5% strain without sacrificing tissue‐like compliance ( E = 127–404 kPa). Using this sensor as human–machine interfaces (HMIs), we facilitate accurate, timely, and stable pressure input for diverse signal waveforms in robotic teleoperation, as well as a deformable, intelligent fingertip for safely detecting soft tissue modulus. Our design provides a promising route to decode sophisticated tactile interactions by linearizing both electrical responses and mechanical behaviors.

Soft robots require the seamless integration of actuation and sensing units to achieve autonomy and adaptability behaviors. This demands conformal, stretchable, and spatially distributed electronic components—an unmet challenge with conventional design and fabrication methods due to complex three-dimensional (3D) geometries, multimaterial integration, and mechanical-electrical mismatches between rigid electronics and soft bodies. Here, we present a design and fabrication framework for autonomous soft robots with embedded actuation and sensing. We develop an integrated digital light processing and direct ink writing 3D printing technology to fabricate soft robots with embedded and conformal 3D electronics in an automated manner. To ensure both electrical and mechanical stability under large deformation, we introduce a structural design strategy incorporating lattice metamaterials, wavy interconnects, and discretized printed circuit boards. By combining the fabrication and design methods, we demonstrate soft robots with multimodal actuation, real-time tactile sensing, wireless communication, tactile-to-visual feedback, and autonomous obstacle avoidance. Our approach paves the way for the development of electronics-integrated autonomous soft robots.

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

Multimaterials with different modulus can endow soft robots with embodied intelligence that deliver spatially varying deformation upon actuation. There is an increasing need for design tools that can rigorously and efficiently generate material layouts for desired motions. Here, we present a design paradigm for soft pneumatic multimaterial actuators by attaching a stiffer material layer as skeleton to softer inflated rubber, and develop a topology optimization based framework to automatically generate the skeleton layout that leads the actuator to achieve desired motions such as bending or twisting. Our method is enabled by a dynamic level set function to describe and track the topological change of the skeleton, large-deformation analysis compatible with the varying skeleton layout, and a gradient-based optimizer to govern the evolution of material layout, with the geometric and material nonlinearities taken into account. A forward geometric mapping and a backward design velocity mapping are constructed to allow manipulating the level sets on the planar space. We show that the design methodology is capable of generating high-performance bending and twisting actuators of cylindrical or customized cone shape. The simulation and experiment results show that, the bending actuator achieves a free bending angle 73° and blocking force 2.05 N, and the twisting actuator achieves a large rotation angle of 143°.

Multichamber soft pneumatic actuators (m‐SPAs) are widely used in soft robotic systems to achieve versatile grasping and locomotion. However, existing m‐SPAs have slow actuation speed and are either limited by a finite air supply or require energy‐consuming hardware to continuously supply compressed air. Herein, these shortcomings by introducing an internal exhaust air recirculation (IEAR) mechanism for high‐speed and low‐energy actuation of m‐SPAs are addressed. This mechanism recirculates the exhaust compressed air and recovers the energy by harnessing the rhythmic actuation of multiple chambers. A theoretical model to guide the analysis of the IEAR mechanism, which agrees well with the experimental results, is developed. Comparative experimental results of several sets of m‐SPAs show that the IEAR mechanism significantly improves the actuation speed by more than 82.4% and reduces the energy consumption per cycle by more than 47.7% under typical conditions. The promising applications of the IEAR mechanism in various pneumatic soft machines and robots such as a robotic fin, fabric‐based finger, and quadruped robot are further demonstrated. An interactive preprint version of the article can be found at: https://doi.org/10.22541/au.166428178.80668101/v1 .

In this paper, we built up a desktop digital light processing (DLP) 3D printer and fabricated multiple size soft pneumatic actuators integrally with fast speed and high precision. The printing process is based on the projection microstereolithography method. The composition of the printing system and key parameters during the printing process were presented. Evaluation experiments demonstrate that our printer can print objects with as small as 87.5 μm size features. We first printed single pneumatic net-works (pneu-net) actuators integrally and conducted a series of actuation experiments and finite element method analyses to test the actuators’ deformation capability. We further designed a soft pneumatic gripper containing three micro pneu-net actuators with 0.4 mm wide square air channels as well as 0.2 mm thick chamber walls and fabricated the gripper integrally in less than 30 min using our printer. The grasping capability of the gripper was verified through experiments as well. Results presented in this work prove the performance of the DLP printer we build up and show the convenience of fabricating micro soft pneumatic actuators integrally using DLP 3D printing approach with fast speed and high precision.

Handheld paradigms offer an efficient and intuitive way for collecting large-scale demonstration of robot manipulation. However, achieving contact-rich bimanual manipulation through these methods remains a pivotal challenge, which is substantially hindered by hardware adaptability and data efficacy. Prior hardware designs remain gripper-specific and often face a trade-off between tracking precision and portability. Furthermore, the lack of online feasibility checking during demonstration leads to poor replayability. More importantly, existing handheld setups struggle to collect interactive recovery data during robot execution, lacking the authentic tactile information necessary for robust policy refinement. To bridge these gaps, we present TAMEn, a tactile-aware manipulation engine for closed-loop data collection in contact-rich tasks. Our system features a cross-morphology wearable interface that enables rapid adaptation across heterogeneous grippers. To balance data quality and environmental diversity, we implement a dual-modal acquisition pipeline: a precision mode leveraging motion capture for high-fidelity demonstrations, and a portable mode utilizing VR-based tracking for in-the-wild acquisition and tactile-visualized recovery teleoperation. Building on this hardware, we unify large-scale tactile pretraining, task-specific bimanual demonstrations, and human-in-the-loop recovery data into a pyramid-structured data regime, enabling closed-loop policy refinement. Experiments show that our feasibility-aware pipeline significantly improves demonstration replayability, and that the proposed visuo-tactile learning framework increases task success rates from 34% to 75% across diverse bimanual manipulation tasks. We further open-source the hardware and dataset to facilitate reproducibility and support research in visuo-tactile manipulation.

Wearable intelligence on human body requires light and soft electronics built on ultrathin supporting films. It would be ideal to have a generic and scalable approach for processing such ultrathin electronics based on various materials to accommodate different design requirements. In this work, a solution based delamination approach is developed using sodium hydroxide (NaOH) treated cross‐linked polymethylmethacrylate (c‐PMMA) film as the separation layer to process ultrathin polymer films. The c‐PMMA layer has resistance to various solvents for coating different polymer materials on top. Reaction of the c‐PMMA surface with mixture of NaOH and water reduces adhesion between the polymer film and the carrier substrate. The polymer film is thus able to be delaminated with facile processes. The approach is applied to fabricate ultrathin flexible or stretchable transparent conductive films, and resistive strain sensors using different polymer materials. A new wrinkle structure capacitive pressure sensor is also realized by integrating different ultrathin polymer films, and presents higher or comparable sensitivity (2.76 kPa −1 ) compared to previous work but with a much lighter weight (<60 g m −2 ). Wearable systems using these sensors are finally built for real time measuring wrist pulse waves, and controlling all actions of a shooting computer game with only three fingers.

No abstract is provided for this article.

Grasping is an important characteristic of robots in interacting with humans and the environment. Due to the inherent compliance of soft grippers, they can easily adapt to novel objects and operate safely in a human-centered environment. However, soft hands suffer from poor grasping robustness and operation durability, especially for heavy objects or objects with sharp spikes, mainly due to their fragile material and low structural stiffness of the soft actuators. Thus, the widespread use of soft hands in daily applications is still limited. Existing works have shown a promising direction to enhance grasping performance by solving the contradiction between inherent compliance/adaptability and loading capacity. It is known that the stiffness of the robotic phalange is highly related to the performance of robotic hands. In this article, we propose a novel variable stiffness particle phalange, called VSPP here. The proposed VSPP exhibits variable stiffness characteristics without the need for dedicated actuation by utilizing passive particle jamming resulted from forces in interacting with the environment. The VSPP can cooperate with any kind of actuators, soft or rigid, to function as a compliant and robust robotic hand. A prototype robotic hand based on VSPP could maintain reliable grasping even when pierced by sharp objects such as a needle, a cactus, and a durian. This durability is effective both in air and underwater, thus presents new possibilities for the soft robotic hand to work in a harsh environment. The inherent multidirectional compliance of the VSPP makes safety in human/robot interaction guaranteed. The design and modeling presented in this research will provide useful guidance in VSPP applications. A prototype gripper, VSPP-3, composed of three 2-segments VSPP fingers and pneumatic joints, has been built for demonstrations in reliable and robust grasping of daily objects. The sample grasping has shown that the proposed VSPP has great potential for a robust and durable soft robotic hand or gripper design.

Handheld paradigms offer an efficient and intuitive way for collecting large-scale demonstration of robot manipulation. However, achieving contact-rich bimanual manipulation through these methods remains a pivotal challenge, which is substantially hindered by hardware adaptability and data efficacy. Prior hardware designs remain gripper-specific and often face a trade-off between tracking precision and portability. Furthermore, the lack of online feasibility checking during demonstration leads to poor replayability. More importantly, existing handheld setups struggle to collect interactive recovery data during robot execution, lacking the authentic tactile information necessary for robust policy refinement. To bridge these gaps, we present TAMEn, a tactile-aware manipulation engine for closed-loop data collection in contact-rich tasks. Our system features a cross-morphology wearable interface that enables rapid adaptation across heterogeneous grippers. To balance data quality and environmental diversity, we implement a dual-modal acquisition pipeline: a precision mode leveraging motion capture for high-fidelity demonstrations, and a portable mode utilizing VR-based tracking for in-the-wild acquisition and tactile-visualized recovery teleoperation. Building on this hardware, we unify large-scale tactile pretraining, task-specific bimanual demonstrations, and human-in-the-loop recovery data into a pyramid-structured data regime, enabling closed-loop policy refinement. Experiments show that our feasibility-aware pipeline significantly improves demonstration replayability, and that the proposed visuo-tactile learning framework increases task success rates from 34% to 75% across diverse bimanual manipulation tasks. We further open-source the hardware and dataset to facilitate reproducibility and support research in visuo-tactile manipulation.

No abstract is provided for this article.

Robotic gloves show promise in hand assistance due to their wearability and home-based potential, yet empirical research remains limited. This pilot study presents a fabric-based pneumatic soft glove, aiming to identify its potential and challenges in clinical practice by evaluating its effectiveness in assisting patients with severe brachial plexus injury (BPI). The glove integrates a thumb abduction actuator and four bidirectional fabric-based pneumatic actuators (FPAs) with asymmetric chambers for high output force. Sixteen healthy volunteers and five individuals with BPI, all of whom lacked active hand and wrist movements, were recruited. Participants performed object grasping across 25 cm. The healthy group performed seven tasks using objects weighing up to 2 kg, with muscle activities recorded for analysis. The BPI group further performed tasks with eight objects from the action research arm test (ARAT) and twelve objects for activities of daily living (ADLs), encompassing various sizes, weights, and geometries. In the healthy group, sEMG showed a decrease in 89.3% of trials, with 56.0% of these decreases being significant (p$< $0.01). For BPI group, the range of motion (ROM) improved, ranging from $28.5 \pm 7.9^{\circ }$ to $63.1 \pm 5.1^{\circ }$ (thumb) and $10.3 \pm 17.5^{\circ }$ to $122.5 \pm 19.0^{\circ }$ (index finger). With a zero baseline for all tasks, their completion rates were $6.8 \pm 0.8$ out of 8 for ARAT tasks and $10.0 \pm 1.7$ out of 12 for ADLs. The fabric-based pneumatic soft glove significantly enhanced the hand function of patients with severe BPI, demonstrating its potential for hand assistance.