Compact Multilayer Extension Actuators for Reconfigurable Soft Robots

4D Multimaterial Printing of Soft Actuators with Spatial and Temporal Control

Soft actuators (SAs) are devices which can interact with delicate objects in a manner not achievable with traditional robotics. While it is possible to design a SA whose actuation is triggered via an external stimulus, the use of a single stimulus creates challenges in the spatial and temporal control of the actuation. Herein, a 4D printed multimaterial soft actuator design (MMSA) whose actuation is only initiated by a combination of triggers (i.e., pH and temperature) is presented. Using 3D printing, a multilayered soft actuator with a hydrophilic pH-sensitive layer, and a hydrophobic magnetic and temperature-responsive shape-memory polymer layer, is designed. The hydrogel responds to environmental pH conditions by swelling or shrinking, while the shape-memory polymer can resist the shape deformation of the hydrogel until triggered by temperature or light. The combination of these stimuli-responsive layers allows for a high level of spatiotemporal control of the actuation. The utility of the 4D MMSA is demonstrated via a series of cargo capture and release experiments, validating its ability to demonstrate active spatiotemporal control. The MMSA concept provides a promising research direction to develop multifunctional soft devices with potential applications in biomedical engineering and environmental engineering.

An Amphibious Fully-Soft Centimeter-Scale Miniature Crawling Robot Powered by Electrohydraulic Fluid Kinetic Energy

Miniature locomotion robots with the ability to navigate confined environments show great promise for a wide range of tasks, including search and rescue operations. Soft miniature locomotion robots, as a burgeoning field, have attracted significant research interest due to their exceptional terrain adaptability and safety features. Here, a fully-soft centimeter-scale miniature crawling robot directly powered by fluid kinetic energy generated by an electrohydraulic actuator is introduced. Through optimization of the operating voltage and design parameters, the average crawling velocity of the robot is dramatically enhanced, reaching 16 mm s-1. The optimized robot weighs 6.3 g and measures 5 cm in length, 5 cm in width, and 6 mm in height. By combining two robots in parallel, the robot can achieve a turning rate of ≈3° s-1. Additionally, by reconfiguring the distribution of electrodes in the electrohydraulic actuator, the robot can achieve 2 degrees-of-freedom translational motion, improving its maneuverability in narrow spaces. Finally, the use of a soft water-proof skin is demonstrated for underwater locomotion and actuation. In comparison with other soft miniature crawling robots, this robot with full softness can achieve relatively high crawling velocity as well as increased robustness and recovery.

An adaptive ensemble deep learning framework for reliable detection of pandemic patients

Nurses, often considered the backbone of global health services, are disproportionately vulnerable to COVID-19 due to their front-line roles. They conduct essential patient tests, including blood pressure, temperature, and complete blood counts. The pandemic-induced loss of nursing staff has resulted in critical shortages. To address this, robotic solutions offer promising avenues. To solve this problem, we developed an ensemble deep learning (DL) model that uses seven different models to detect patients. Detected images are then used as input for the soft robot, which performs basic assessment tests. In this study, we introduce a deep learning-based approach for nursing soft robots, and propose a novel deep learning model named Deep Ensemble of Adaptive Architectures. Our method is twofold: firstly, an ensemble deep learning technique detects COVID-19 patients; secondly, a soft robot performs basic assessment tests on the identified patients. We evaluate the performance of various deep learning-based object detectors for patient detection, examining implementations of You Only Look Once (YOLO), Single Shot MultiBox Detector (SSD), Region-Based Convolutional Neural Network (RCNN), and Region-Based Fully Convolutional Network (R-FCN) on a proprietary dataset comprising 32,668 hospital surveillance images. Our results indicate that while YOLO and VGG facilitate rapid detection, Faster-RCNN (Inception ResNet-v2) and our proposed Ensemble-DL achieve the highest accuracy. Ensemble-DL offers accurate results in a reasonable timeframe, making it apt for patient detection on embedded platforms. Through real-world experiments, our method outperforms baseline approaches (including Faster-RCNN, R-FCN variants, CNN+LSTM, etc.) in terms of both precision and recall. Achieving an impressive accuracy of 98.32%, our deep learning-based model for nursing soft robots presents a significant advancement in the identification and assessment of COVID-19 patients, ultimately enhancing healthcare efficiency and patient care.

Reconfigurable Soft Pneumatic Actuators Using Extensible Fabric-Based Skins

The development of the field of soft robotics has led to the exploration of novel techniques to manufacture soft actuators, which provide distinct advantages for wearable assistive robotics. One subset of these soft pneumatic actuators is conventionally developed from silicone, fabrics, and thermoplastic polyurethane (TPU). Each of these materials in isolation possesses limitations of low-stress capacity, low-design complexity, and high-input pressure requirements, respectively. Combining these materials can overcome some limitations and maintain their desirable properties. In this article, we explore one such composite design scheme using a combination of silicone polymer-based bladder and reconfigurable fabric skin made from an anisotropic extensible fabric. The silicone polymer bladder acts as the hermetic seal, while this skin acts as the constraint. Bending and torsional actuators were designed utilizing the anisotropy of these fabrics. The torsional actuator designs can achieve over 540° of twist, significantly larger than previously reported in the literature, owing to the lower mechanical impedance of the extensible fabrics. Actuators with 360° of bending were also fabricated using this method. In addition, the lack of TPU-backed or inextensible fabrics reduces the actuator's stiffness, leading to lower actuation pressures. Skin-based designs also confer the advantage of modularity, reconfigurability, and the ability to achieve complex motions by tuning the properties of the bladder and the skin. For applications with high-force requirements, such as wearable exoskeletons, we demonstrate the utility of multilayer design schemes. A multilayer bending actuator generated 190 N of force at 100 kPa and was shown to be a candidate for wearable assistive devices. In addition, torsional designs were shown to have utility in practical scenarios such as screwing on a bottle cap and turning knobs. Thus, we present a novel fabric-skin-based design concept that is highly versatile and customizable for various application requirements.