Soft robots can accomplish hand rehabilitation training to ensure better safety and compliance for hand rehabilitation. In this study, a wavy nonrotating soft actuator structure was proposed for hand rehabilitation, and an axial stiffener was added to the main structure of the actuator according to the function of the bamboo fiber. A physical model of the actuator was fabricated using a multistep casting molding method, and the performance of the designed soft actuator was tested experimentally. The results showed that the bending angle and contact force gradually increased with increasing pressure. The average maximum bending angle and contact force can reach 286 ± 14.3 deg and 1.04 ± 0.051 N, with a pressure of 72 kPa. Meanwhile, the bending torques of the soft actuator at each joint of the finger were tested, to verify that it can meet the needs of soft actuators for hand applications. Furthermore, the load lifting of the soft actuator with axial stiffeners can increase by 6 mm on average compared with a soft actuator without axial stiffeners under negative pressure. In conclusion, the pneumatic soft actuator can produce two different motion functions under the action of one cavity. In addition, a soft actuator with an axial stiffener can improve the load capacity under negative pressure. By assembling the actuators, a three-finger gripper was manufactured. The gripper could grasp and lift objects. Therefore, this work provides a new route for the development of pneumatic soft actuators and soft robots, which has efficient driving.

Frequent repetitive manual tasks in the manufacturing industry have contributed to a rise in hand motor dysfunction, negatively affecting both health and work efficiency. While rigid hand exoskeleton robots are effective, they are often heavy and may cause secondary injuries. In contrast, soft hand exoskeleton robots demonstrate significant potential as tools for injury prevention and assistance. This paper presents a hand-wearable pneumatic soft robot designed to assist in daily hand activities and prevent muscle atrophy. The developed soft actuator features a nested structure with uniformly distributed cross-sectional shapes of the internal pneumatic chamber. A constraining shell, made of materials with a different elastic modulus than the main structure, is placed around the pneumatic chamber to limit radial expansion. Based on the general structural parameters of the cross-sectional shape, a physical model was established to simulate bending conditions. The actuator prototype was fabricated using a multistep casting method. During experiments, at an air pressure of 43 kPa, the actuator achieved an average bending angle of 260 ± 5° and an average output force of 6.0 ± 0.1 N. The soft actuator was integrated into a wearable hand robot for further testing. Results demonstrate that it effectively supports finger flexion and object grasping while helping to reduce muscle fatigue.

In order to achieve controllable behavior in soft robots, it is necessary to analyze the dynamic response characteristics of soft actuators and thereby gain a deeper understanding of the dynamic response mechanisms of actuators. In this study, the two-way fluid structural interaction (FSI) method was used for the first time to reveal the transient flow field inside the soft pneumatic actuator and its influences on the dynamic motions under high-speed pressurization. Moreover, the simulation results were verified by experimentally by using a high-speed camera to capture the transient morphology of the actuator. The simulation results indicated that streamlines inside the actuator were not uniformly distributed, and the air pressure was not even either. Driven by the high-speed air flow, the head of actuator bent at first, resulting in the expansion of the chambers gradually, followed by the arching of the actuator and more violent variation in air streamlines and pressure distribution. Accompanying the deformation of the actuator, the stress distribution, as well as the streamline and pressure, changed dramatically within a total 0.1 second. Experimentally, the transient morphologies captured by the high-speed camera at various moments coincided with the simulation results. Compared to the conventional solid-mechanics finite element analysis (FEA) method, the FSI simulation showed better precision and accuracy. This work serves as a valuable reference for research on the fine-structured design and controllable motion of soft actuators and soft robots

Soft robots demonstrate significant advantages in applications within complex environments due to their unique material properties and structural designs. However, they also face challenges in fault diagnosis, such as nonlinearity, time variability, and the difficulty of precise modeling. To address these issues, this paper proposes a fault diagnosis method based on multimodal spatiotemporal features and ensemble learning. First, a sliding-window Kalman filter is utilized to eliminate noise interference from multi-source signals, constructing separate temporal and spatial representation spaces. Subsequently, an adaptive weight strategy for feature fusion is applied to train a heterogeneous decision tree model, followed by a dynamic weighted voting mechanism based on confidence levels to obtain diagnostic results. This method optimizes the feature extraction and fusion process in stages, combined with a dynamic ensemble strategy. Experimental results indicate a significant improvement in diagnostic accuracy and model robustness, achieving precise identification of faults in soft robots.