This letter introduces a novel bistable origami flexible gripper, which is based on a single-vertex and multi-crease (SVMC) origami structure that has several advantages, including a simple structure, low cost, and strong deformation capacity. This design addresses the drawbacks of slow response speed and low grasping efficiency in traditional gripper models. The origami structure is established using a waterbomb model, and the relationship between the crease angles is analyzed by applying the geometric relationship of spherical triangle to establish kinematic equations. In addition, potential energy equations in the deformation of SVMC are derived using a torsion spring model, which parameters are determined by conducting a detailed analysis of the influence of crease length and initial angle on energy storage and release processes. In addition, the tendon drive system helps to provide greater grasping force, promoting efficient and stable target grasping when the SVMC return to the initial state. Experimental results show that when subjected to an external trigger force of 2.65 N, the flexible gripper can transition from a spreading attitude to a contracted state within 61.58 ms, enabling rapid envelopment of the target, dropping the target, and returning to the initial state within 120 ms. These findings suggest that the proposed flexible gripper has significant application in high-speed food sorting and fruit picking.

The Fundus photography vs Ophthalmoscopy Trial Outcomes in the Emergency Department (FOTO-ED) studies showed that ED providers poorly recognized funduscopic findings in patients in the ED. We tested a modified version of the Brain and Optic Nerve Study Artificial Intelligence (BONSAI) deep learning system on nonmydriatic fundus photographs from the FOTO-ED studies to determine if the deep learning system could have improved the detection of papilledema had it been available to ED providers as a real-time diagnostic aid. Retrospective secondary analysis of a cohort of patients included in the FOTO-ED studies. The testing data set included 1608 photographs obtained from 828 patients in the FOTO-ED studies. Photographs were reclassified according to the optic disc classification system used by the deep learning system ("normal optic discs," "papilledema," and "other optic disc abnormalities"). The system's performance was evaluated by calculating the area under the receiver operating characteristic curve (AUC), sensitivity, and specificity using a 1-vs-rest strategy, with reference to expert neuro-ophthalmologists. The BONSAI deep learning system successfully distinguished normal from abnormal optic discs (AUC 0.92 [95% confidence interval {CI} 0.90-0.93]; sensitivity 75.6% [73.7%-77.5%] and specificity 89.6% [86.3%-92.8%]), and papilledema from normal and others (AUC 0.97 [0.95-0.99]; sensitivity 84.0% [75.0%-92.6%] and specificity 98.9% [98.5%-99.4%]). Six patients with missed papilledema in 1 eye were correctly identified by the deep learning system as having papilledema in the other eye. The BONSAI deep learning system was able to reliably identify papilledema and normal optic discs on nonmydriatic photographs obtained in the FOTO-ED studies. Our deep learning system has excellent potential as a diagnostic aid in EDs and non-ophthalmology clinics equipped with nonmydriatic fundus cameras. NOTE: Publication of this article is sponsored by the American Ophthalmological Society.

Delicate underwater manipulation tasks such as biological specimen collection are promising fields that require new robotic designs and intelligent robotic technologies. In this study, we proposed an automatic aquatic object-collecting system with a soft manipulator controlled by a reinforcement learning-based controller. For underwater sensing, we implemented a visual perception framework to restore the quality of the underwater image, detect the seafood animals, and track the target's position. The online learning ability of the reinforcement learning-based controller endowed strong adaptability for the soft manipulator against underwater disturbances. The water tank grasping tests show a 91.7% successful grasping rate without flow disturbance and 83.3% with flow disturbances. We demonstrated that the soft robotic collecting system gripped seafood animals in a lab aquarium as well as the natural seabed environment. The real-world experimental results showed that the robot successfully collected 28 shells within 40 min at a water depth of 15 m and even completed grasping tasks in a dark environment. Our results demonstrated that this manipulator prototype is potentially applicable for fully autonomous delicate objects underwater.

This article investigates the problem of model-free tracking control for a class of continuum manipulators. A new type of robust control strategy is provided to address the problem, which primarily consists of two parts. First, to achieve output tracking with the desired accuracy, an error transformation scheme is constructed. Moreover, because of the robustness of the error transformation scheme against model uncertainties, no knowledge of the manipulator model is required, leading to a model-free control solution. Then, a tuning function is used to modify certain error variables to relax the requirements on the initial conditions of the error transformation scheme. In this way, the global stability of the closed-loop system can be guaranteed. Finally, both simulation and experimental results validate the expected performance of the presented control method.

The outbreak of COVID-19 has caused immeasurable loss to the world. The highly contagious virus poses a challenging risk to public health, especially medical doctors, who are in close contact with patients. Thus, robots can help reduce the risk of medical professionals conducting massive testing with throat swabs. Soft robots, fabricated with compliant materials, allow for low-cost development and safe interaction with humans and show promising potentiality in testing.