Journal of Bionic Engineering最新文献

At present, the existing piezoelectric stick-slip actuators have an inherent back-slip problem, which greatly limits the development and application of stick-slip actuators. In order to inhibit the regression phenomenon, a new bionic lemongrass stickslip actuator was prepared by using polymer PDMS to replicate natural biological surface. The surface microstructure of the grass was copied by PDMS, and the PDMS film was prepared. The rigid and flexible bionic friction pair was further prepared, and the flexible anisotropic PDMS stick slip actuator was developed. It was found that the anisotropic friction characteristics of the surface microstructure of the grass inhibited the anti-sliding motion, and the elastic potential energy of the PDMS film improved the output characteristics of the driver. By adjusting the input voltage to control the contact between the drive foot and the rotor, the rigid and flexible hybrid drive can be realized and the backsliding phenomenon can be suppressed. The actuator is compact, lightweight and can achieve high speed and high resolution output without preloading force, which has important application value in the field of fast and accurate positioning with load limitation.

Butterfly coloration originates from the finely structured scales grown on the underlying wing cuticle. Most researchers who study butterfly scales are focused on the static optic properties of cover scales, with few works referring to dynamic optical properties of the scales. Here, the dynamic coloration effect of the multiple scales was studied based on the measurements of varying-angle reflection and the characterization of scale flexibility in two species of Lycaenid, Plebejus argyrognomon with violet wings and Polyommatus erotides with blue wings. We explored the angle-dependent color changeability and the color-mediating efficiency of wing scales. It was found that the three main kinds of flexible scales (cover, ground and androconia scales) were asynchronously bent during wing rotation, which caused the discoloration effect. The three layers of composite scales broaden the light signal when compared to the single scale, which may be of great significance to the recognition of insects. Specifically, the androconia scales were shown to strongly contribute to the overall wing coloration. The cover scale coloration was ascribed to the coherence scattering resulted from the short-range order at intermediate spatial frequencies from the 2D Fourier power spectra. Our findings are expected to deepen the understanding of the complex characteristics of biological coloration and to provide new inspirations for the fabrication of biomimetic flexible discoloration materials.

The transistor-inspired Droplet-based Electricity Generator (DEG) significantly enhances the energy collection efficiency from single-position droplets. However, the design of the DEG arrays combining high output performance and large-scale integration under multi-position droplet impacts remains a challenge. Inspired by the unique structure of the honeycomb, we developed an Independent-Cell Droplet-based Electricity Generator (IC-DEG) array that allows for high-efficiency and stable droplet energy harvesting under multi-position droplet impacts. Each independent cell is a transistor-inspired Tubular Droplet-based Electricity Generator (T-DEG), which ensures the high electrical output of the IC-DEG array. The honeycomb-like arrangement improves the space utilization, accelerates the detachment of droplets, and avoids electrical interference among independent cells, all of which further enhance the IC-DEG array performance. The average peak open-circuit voltage of the IC-DEG array is 265.2 V, and 96.6% of peak voltages exceed 200 V, almost double that of a traditional planar array. Moreover, the average droplet detachment time of the IC-DEG array is 44.8 ms, 41.4% shorter than the traditional planar array. The enhanced performance of the IC-DEG array is further demonstrated by the high speed of charging capacitors and the capability of driving electronic devices. This study provides a promising design concept for large-scale droplet energy harvesting devices.

Sustainable cement-based concrete materials are primarily used for construction, among which vermiculite as lightweight fine aggregate gains more future development prospect. First, a bacterial solution was sprayed over vermiculite and wrapped using calcium sulphoaluminate (CSA) cement to replace with fine aggregate in concrete. Secondly, based on a preliminary test on compressive strength results, 10% of Ground Granulated Blast Furnace Slag (GGBS) and a healing solution proportion of 9:1 was selected for preparing self-healing concrete. The fine aggregate was replaced in concrete using vermiculite in 0%, 5%, 10% and 15% and the findings suggest that bacterial vermiculite replacement should be at most 5% to achieve better results in strength and durable properties. The strength enhancement observed for compressive strength, strength regain, split tensile strength, flexural strength, and ultrasonic pulse velocity were 29.22%, 45.5%, 34.02%, 28.03% and 41.4% respectively. Surface crack healing at 7, 14 and 28 days of BIVC was 38.23%, 58.82% and 79.41%, which is 3–4% lower than internal crack healing. Microstructural analysis by Scanning Electron Microscopy (SEM), X-Ray Diffractometer (XRD), and Energy Dispersive Spectroscopy (EDS) reveals the existence of calcite, and it was formed due to the bio-mineral action of bacteria with available nutrients in sustainable concrete.

Automatic identification and segmentation of lesions in medical images has become a focus area for researchers. Segmentation for medical image provides professionals with a clearer and more detailed view by accurately identifying and isolating specific tissues, organs, or lesions from complex medical images, which is crucial for early diagnosis of diseases, treatment planning, and efficacy tracking. This paper introduces a deep network based on dendritic learning and missing region detection (DMNet), a new approach to medical image segmentation. DMNet combines a dendritic neuron model (DNM) with an improved SegNet framework to improve segmentation accuracy, especially in challenging tasks such as breast lesion and COVID-19 CT scan analysis. This work provides a new approach to medical image segmentation and confirms its effectiveness. Experiments have demonstrated that DMNet outperforms classic and latest methods in various performance metrics, proving its effectiveness and stability in medical image segmentation tasks.

The In Vitro Bionic Digestion Model (IVBDM) are used to simulate the digestion process of food or pharmaceuticals in corresponding digestion tracts for obtaining the digestion data, which are expected to replace in vivo experiments with animals in the early stages of functional food or drug development, and thus have broad applications prospects. However, little is known so far about how the factors including the Young’s modulus of the model, the level, location and direction of the applied load, affect the peristalsis amplitude of the IVBDM. Based on an In Vitro Bionic Rat Stomach Model (IVBRSM), simulation and experimental analysis were conducted to examine the factors effecting the peristalsis amplitude of the IVBRSM. It is shown that Young’s modulus of the model significantly affects the peristalsis amplitude, with lower Young’s modulus resulting in larger amplitude. Load level, location, and direction also influence the peristalsis amplitude. Additionally, IVBRSM size and wall thickness play a role, with larger models requiring higher load levels or lower Young’s modulus for the same peristalsis amplitude. Simulation data correlate well with experimental results. These findings contribute to the understanding of the peristalsis state of IVBRSM under different conditions and can guide the design and fabrication of such in vitro bionic digestion models.

Effectively controlling active power-assist lower-limb exoskeletons in a human-in-the-loop manner poses a substantial challenge, demanding an approach that ensures wearer autonomy while seamlessly adapting to diverse wearer needs. This paper introduces a novel hierarchical control scheme comprising five integral components: intention recognition layer, dynamics feedforward layer, force distribution layer, feedback compensation layer, as well as sensors and actuators. The intention recognition layer predicts the wearer’s movement and enables wearer-dominant movement through integrated force and position sensors. The force distribution layer effectively resolves the statically indeterminate problem in the context of double-foot support, showcasing flexible control modes. The dynamics feedforward layer mitigates the effect of the exoskeleton itself on movement. Meanwhile, the feedback compensation layer provides reliable closed-loop control. This approach mitigates abrupt changes in joint torques during frequent transitions between swing and stance phases by decomposed dynamics. Validating this innovative hierarchical control scheme on a hydraulic exoskeleton platform through a series of experiments, the results demonstrate its capability to deliver assistance in various modes such as stepping, squatting, and jumping while adapting seamlessly to different terrains.

With the increasing dimensionality of the data, High-dimensional Feature Selection (HFS) becomes an increasingly difficult task. It is not simple to find the best subset of features due to the breadth of the search space and the intricacy of the interactions between features. Many of the Feature Selection (FS) approaches now in use for these problems perform significantly less well when faced with such intricate situations involving high-dimensional search spaces. It is demonstrated that meta-heuristic algorithms can provide sub-optimal results in an acceptable amount of time. This paper presents a new binary Boosted version of the Spider Wasp Optimizer (BSWO) called Binary Boosted SWO (BBSWO), which combines a number of successful and promising strategies, in order to deal with HFS. The shortcomings of the original BSWO, including early convergence, settling into local optimums, limited exploration and exploitation, and lack of population diversity, were addressed by the proposal of this new variant of SWO. The concept of chaos optimization is introduced in BSWO, where initialization is consistently produced by utilizing the properties of sine chaos mapping. A new convergence parameter was then incorporated into BSWO to achieve a promising balance between exploration and exploitation. Multiple exploration mechanisms were then applied in conjunction with several exploitation strategies to effectively enrich the search process of BSWO within the search space. Finally, quantum-based optimization was added to enhance the diversity of the search agents in BSWO. The proposed BBSWO not only offers the most suitable subset of features located, but it also lessens the data’s redundancy structure. BBSWO was evaluated using the k-Nearest Neighbor (k-NN) classifier on 23 HFS problems from the biomedical domain taken from the UCI repository. The results were compared with those of traditional BSWO and other well-known meta-heuristics-based FS. The findings indicate that, in comparison to other competing techniques, the proposed BBSWO can, on average, identify the least significant subsets of features with efficient classification accuracy of the k-NN classifier.

Power systems are pivotal in providing sustainable energy across various sectors. However, optimizing their performance to meet modern demands remains a significant challenge. This paper introduces an innovative strategy to improve the optimization of PID controllers within nonlinear oscillatory Automatic Generation Control (AGC) systems, essential for the stability of power systems. Our approach aims to reduce the integrated time squared error, the integrated time absolute error, and the rate of change in deviation, facilitating faster convergence, diminished overshoot, and decreased oscillations. By incorporating the spiral model from the Whale Optimization Algorithm (WOA) into the Multi-Objective Marine Predator Algorithm (MOMPA), our method effectively broadens the diversity of solution sets and finely tunes the balance between exploration and exploitation strategies. Furthermore, the QQSMOMPA framework integrates quasi-oppositional learning and Q-learning to overcome local optima, thereby generating optimal Pareto solutions. When applied to nonlinear AGC systems featuring governor dead zones, the PID controllers optimized by QQSMOMPA not only achieve 14(%) reduction in the frequency settling time but also exhibit robustness against uncertainties in load disturbance inputs.

The development of a tailless Flapping Wing Micro Aerial Vehicle (FWMAV) inspired by the hummingbird is presented in this work. By implementing mechanical simplifications, it is possible to use planar machining technology for manufacturing of the FWMAV’s body, greatly reducing assembly errors. Traditionally, studies on flapping wing aircraft are limited to open-loop wing kinematics control. In this work, an instantaneous closed-loop wing trajectory tracking control system is introduced to minimize wings’ trajectory tracking errors. The control system is based on Field-Oriented Control (FOC) with a loop shaping compensation technique near the flapping frequency. Through frequency analysis, the loop shaping compensator ensures the satisfactory bandwidth and performance for the closed-loop flapping system. To implement the proposed controller, a compact autopilot board integrated with FOC hardware is designed, weighing only 2.5 g. By utilizing precise wing trajectory tracking control, the hummingbird-inspired FWMAV demonstrates superior ability to resist external disturbances and exhibits reduced attitude tracking errors during hovering flight compared to the open-loop wing motion.