Achieving multimodal kinematic paths (i.e. having more than one path) in adaptive structures often requires active manipulation, such as actuation or locking, which increases system complexity. This paper addresses the gap in naturally engaging and disengaging constraint with external loads through tendon constrained inflatables (TCIs), without reliance on active control. A TCI consists of an inflatable bladder and rigid end caps, connected by tendons. Upon pressurization, these tendons become taut, constraining select degrees of freedom (DOFs). The movement of one end cap relative to the other depends on the remaining unconstrained DOFs. By strategically designing the tendon configuration, a TCI can reconfigure its constraints to selectively allow or constrain motions based on the applied external loads without actuation, sensing, and control as different tendons become taut and slack. This paper introduces load-activated multimodal tendon constrained inflatables that can adaptively provide spatially sequential 1 DOF motion segments by traversing between segment points (i.e. fully constrained positions). A 3D kineto-static equilibrium model is proposed to predict the loads needed to slacken specific tendons and define the TCI’s workspace. Optimization is applied to design tendon configurations and a prototype TCI was fabricated and experimentally validated.
Electromechanical impedance (EMI) technology is a non-destructive method used for structural health monitoring (SHM), with the advantage of being easy to implement and sensitive to early-stage damage. However, the reliability of EMI technology is significantly affected by environmental vibration, as the coupling between lead zirconate titanate (PZT) and structures can be casually disrupted by environmental vibration, leading to false alarm in SHM. To address this problem, based on the fact that the effect of environmental vibration on damage monitoring index is random and the effect of structural damage on damage monitoring index is irreversible, this paper proposes an improved EMI algorithm that combines the statistical principle and the reward-penalty mechanism (RPM). Experiments were conducted to identify the damage of an aluminum alloy cantilever beam in strong vibration environment, and the damage of the cantilever beam was successfully identified after several rounds of SHM. The experimental results confirmed the effectiveness of the improved EMI algorithm and extended the range of applications for EMI technology.
Conventional positioning systems for non-destructive testing (NDT) are often bulky, inaccurate, and constrained in motion range, making them unsuitable for inspecting small, intricate components like chips or for long-distance scans. To overcome these limitations, we propose a novel miniature traveling-wave piezoelectric actuator with high positioning performance. The actuator features a simple cross-shaped structure, driven by dual piezoelectric elements to generate a traveling wave. Characterization demonstrates a no-load speed of 350 mm/s, a load capacity (>50 g) exceeding seven times its own weight, excellent linearity (close to 1), repeatability (>90%), unlimited stroke, and high motion resolution (0.023 mm). When integrated with an ultrasonic NDT probe, the system successfully performed both 2D and 3D scanning tasks, overcoming the constraints of conventional systems. This work extends the application of piezoelectric actuators to NDT and provides a viable pathway toward miniaturized, high-performance inspection systems.
McKibben Artificial Muscles (AMs) are widely used in soft robotics as actuators due to their compliance and high force output. However, achieving complex motions with multiple AMs in different orientations remains challenging, as each requires a bulky pneumatic or hydraulic system, complicating assembly and increasing the risk of leaks. These limitations restrict their use in compact or lightweight applications. Inspired by natural muscle architectures that optimize force distribution and enable multidirectional motion, this study aims to integrate complex fiber arrangements into a single McKibben AM. A novel fabrication technique was developed using an interplay between multiple traditional Japanese Kumihimo braiding disks to create multi-directional AMs. Two case studies demonstrate the approach: (1) a bending actuator capable of achieving programed bending angles depending on actuation pressure and (2) a convergent actuator whose combined branches reconfigured output force direction to extend motion range. The resulting branched AMs locally encode complex motion within their structure, eliminating the need for complex actuator systems. This manufacturing strategy advances McKibben AM design toward more compact, adaptive and versatile systems for advanced soft robotic applications.
As the strong nonlinearity and highly complex of piezoelectric energy harvesters (PEH), accurately predicting the dynamic response using mathematical models remains challenging. This study introduces a CNN-LSTM (Convolutional Neural Network, Long Short Term Memory) combined with a neural optimization machine (NOM) to predict the nonlinear dynamics of the PEH. CNN extracts local patterns from inputs related to incentives, while LSTM layer captures their temporal evolution. NOM uses a differentiable surrogate model to automatically optimize the hyperparameters of the CNN-LSTM, thereby improving the predicting accuracy. The proposed framework avoids explicit dynamic modeling, computationally numerical simulation, and repetitive parameter identification. The framework was then applied to predict the dynamic responses (including tip displacement, adaptive rotation angle and output voltage) of a nonlinear, directional self-adaptive PEH, and the results showed that the determination coefficients ( R 2 ) of all three responses exceeded 0.998, and the root mean square errors (RMSE) of tip displacement, adaptive rotation angle, and output voltage were 0.1156 mm, 0.2736°, and 0.2313 V respectively. There was good consistency between the predicted and measured responses, validating the effectiveness of the proposed method and opening up opportunities for accurately predicting the dynamic characteristics of highly complex and/or strongly nonlinear energy harvesting systems.
