Biomimetics最新文献

The Red-Billed Blue Magpie Optimizer (RBMO) is a recently introduced swarm-based meta-heuristic that has shown strong potential in engineering optimization but remains under-explored. To address its inherent limitations, this paper proposes an Enhanced RBMO (ERBMO) that synergistically incorporates two key strategies: a dominant-group-based two-stage covariance-driven strategy that captures evolutionary trends to improve population quality while reinforcing global exploration, and a Powell mechanism (PM) that eliminates dimensional stagnation and markedly strengthens convergence. Extensive experiments on the CEC 2017 benchmark suite demonstrate that ERBMO outperforms ten basic and improved algorithms in global exploration, local convergence accuracy and robustness, attaining Friedman ranks of 1.931, 1.621, 1.345 and 1.276 at 10D, 30D, 50D and 100D, respectively. Furthermore, empirical studies on practical engineering design problems confirm the algorithm's capability to consistently deliver high-quality solutions, highlighting its broad applicability to real-world constrained optimization tasks. In future work, we will deploy the algorithm for real-world tasks such as UAV path-planning and resource-scheduling problems.

To investigate the impact of flexible versus rigid bioinspired flapping hydrofoils on the discharge characteristics of suspended particles in raceway aquaculture, this study established a two-way fluid-structure coupling model of a flapping hydrofoil device based on ANSYS Fluent and Transient Structural modules. The research compares the discharge characteristics of hydrofoils with different elastic moduli. The results show that, within a certain range of elastic moduli adjustment, flexible bioinspired hydrofoils exhibit greater surface deformation compared to rigid ones, effectively delaying tail vortex shedding and extending its duration, thus prolonging the range of high flow velocities. During the middle stage of discharge, the escape rate of suspended particles under the influence of flexible bioinspired hydrofoils with 0.05 GPa elastic modulus was 3-4% higher than that of rigid hydrofoils. However, in terms of achieving maximum discharge efficiency and effectiveness, both reached approximately 97.8% with little difference between them. This study highlights the bioinspired principles in hydrofoil design and provides a reference for optimizing flexible hydrofoil discharge characteristics in future research.

Defect-rate uncertainty creates cascading operational challenges in multi-stage production, often driving inefficiency and misallocation of labor, materials, and capacity. To confront this, we develop a multi-stage Production Integrated Decision (MsPID) framework that unifies quality inspection and shop-floor decision-making within a single computational model. The framework couples a two-stage sampling inspection policy-used to statistically learn and control defect-rate uncertainty via estimation and rejection rules-with a multi-process, multi-part production decision model. Optimization is carried out with an Improved Grey Wolf Optimizer (IGWO) enhanced with Latin hypercube sampling (LHS) for uniformly diverse initialization; an evolutionary factor mechanism that blends simulated binary crossover (SBX) among three leadership-guided parents (Alpha, Beta, Delta) to strengthen global exploration in early iterations and focus exploitation later; and a greedy, mutation-assisted opposition learning step applied to the lowest-performing quartile of the population to effect leader-informed local refinement and accept only fitness-improving moves. Experiments show the method identifies minimum-cost policies across six single-stage benchmark cases and yields a total profit of 43,800 units in a representative multi-stage scenario, demonstrating strong performance in uncertain environments. Sensitivity analysis further clarifies how recommended decisions adapt to shifts in estimated defect rates, finished product prices, and swap/changeover losses. These results highlight how bio-inspired intelligence can enable adaptive, efficient, and resilient integrated production management at scale.

Background: Developing reliable and cost-effective decellularization methods is critical for advancing tissue engineering and regenerative medicine, particularly in regions with limited access to specialized perfusion systems.

Methods: This study standardized a gravity-assisted perfusion protocol for rat liver decellularization, designed to operate without pumps or pressurized equipment. Adult Wistar rat livers were processed through a gravity-driven vascular flushing method and compared with a conventional immersion-based protocol. The resulting scaffolds were evaluated by macroscopic inspection, histological staining (Masson's trichrome), and residual DNA quantification.

Results: The gravity-assisted perfusion method achieved more efficient cellular removal and superior preservation of extracellular matrix (ECM) integrity compared with immersion. Residual DNA levels were 3.7 ng/mg in perfused samples, 209.47 ng/mg in immersed samples, and 331.97 ng/mg in controls, confirming a statistically significant reduction (p < 0.05). Only the perfused group met the accepted threshold for effective decellularization (<50 ng/mg dry tissue). Histological analysis corroborated these findings, showing the absence of nuclei and the preservation of collagen architecture characteristic of a structurally intact ECM.

Conclusions: This low-cost, reproducible, and technically simple system enables the generation of high-quality acellular hepatic scaffolds without mechanical pumps. Its accessibility and scalability make it suitable for laboratories with limited infrastructure and educational settings. Moreover, this gravity-assisted approach provides a foundation for future recellularization and preclinical studies aimed at developing bioengineered liver constructs for regenerative and transplant applications.

Background: Quantitatively detecting whether plants exhibit measurable bioelectric differences in the presence of nearby human movement remains challenging, in part because plant signals are low-amplitude, slow, and easily confounded by environmental factors. Methods: We recorded bioelectric activity from 2978 plant samples across three species (basil, salad, tomato) using differential electrode pairs (leaf and soil electrodes) sampling at 142 Hz. Two trained performers executed three specific eurythmic gestures near experimental plants while control plants remained isolated. Random Forest and Convolutional Neural Network classifiers were applied to distinguish the control from treatment conditions using engineered features including spectral, temporal, wavelet, and frequency domain characteristics. Results: Random Forest classification achieved 62.7% accuracy (AUC = 0.67) distinguishing differences in recordings collected near a moving human from control conditions, representing a statistically significant 12.7 percentage point improvement over chance. Individual performer signatures were detectable with 68.2% accuracy, while plant species classification achieved only 44.5% accuracy, indicating minimal species-specific artifacts. Temporal analysis revealed that the plants with repeated exposure exhibited consistently less negative bioelectric amplitudes compared to single-exposure plants. Innovation: We introduce a data-driven approach that pairs standardized, short-window bioelectric recordings with machine-learning classifiers (Random Forest, CNN) to test, in an exploratory manner, whether plant signals differ between human-moving-nearby and isolation conditions. Conclusions: Plants exhibit modest but statistically detectable bioelectric differences in the presence of nearby human movement. Rather than attributing these differences to eurythmic movement itself, the present design can only demonstrate that plant recordings collected within ~1 m of a moving human differ, modestly but statistically, from recordings taken ≥3 m away. The underlying biophysical pathways and specific contributing factors (airflow, VOCs, thermal plumes, vibration, electromagnetic fields) remain unknown. These results should therefore be interpreted as exploratory correlations, not mechanistic evidence of gesture-specific plant sensing.

This study aimed to evaluate the biomechanics of maxillary first molar teeth following palatal, disto-buccal, and mesio-buccal root amputation. An intact maxillary molar underwent root canal treatment using Reciproc R25 files (VDW, Munich, Germany). The canals were obturated with gutta-percha (DiaDent, Seoul, Republic of Korea) and 2Seal sealer (VDW, Munich, Germany), and the access cavity was restored with composite resin. A high-resolution CBCT scan of an intact maxillary first molar was obtained using a Planmeca Promax 3D Max system (Planmeca Oy, Helsinki, Finland) at 75 kVp and 10 mA. The acquired data were processed in 3D Slicer software (v5.8.0, BSD license, Boston, MA, USA) to segment enamel, dentin, and pulp based on pixel density variations using the three-point cloud method. A baseline intact model and three root-resected models (palatal, disto-buccal, mesio-buccal) were reconstructed in SolidWorks 2021, with resected roots simulated as being sealed with MTA. Finite element analysis was conducted in CosmosWorks to evaluate von Mises stress distribution under a 300 N static occlusal load. Maximum von Mises stresses were detected at occlusal force application sites. Among root dentin tissues, stress values ranked highest after palatal root resection, followed by the mesio-buccal, disto-buccal, and non-resected models. Conclusions: Palatal root amputation of maxillary first molars generated the highest von Mises stresses in root dentin, suggesting a higher biomechanical risk than disto-buccal or mesio-buccal resections.

There is a growing demand for underwater robots to support offshore tasks such as exploration, environmental monitoring, and critical underwater missions. To enhance the performance of these systems, researchers are increasingly turning to biological inspiration to develop robots that understand and adapt the swimming strategies of aquatic animals. Numerical modeling plays a critical role in evaluating and improving the performance of these complex, multi-body robotic systems. However, developing accurate models for multi-body robots that swim freely in three dimensions remains a significant challenge. This study presents the development and validation of a numerical model of a bio-inspired California sea lion (Zalophus californianus) robot. The model was developed to simulate, analyze, and visualize the robot's body motions in water. The equations of motion were derived in closed form using the Euler-Poincaré formulation, offering advantages for control and stability analysis. Hydrodynamic coefficients essential for estimating fluid forces were computed using computational fluid dynamics (CFD) and strip theory and further refined using a genetic algorithm to reduce the sim-to-real gap. The model demonstrated strong agreement with experiments, accurately predicting the translation and orientation of the robot. This framework provides a validated foundation for simulation, control, and optimization of bio-inspired multi-body systems.

Cactaceae are plants equipped with spines and adapted to extremely arid environments. In particular, Echinocactus grusonii spines are almost cylindrical structures, which may occasionally present an enlargement of their proximal cross sectional area. In this work, the spines of Echinocactus grusonii were explored as a possible source of biomimetic inspiration for the design and the implantation strategies of self-inserting intraneural interfaces. More specifically, the elastic stability of spines was theoretically studied for structures able to puncture the surface of an external object, as well as for structures unable to pierce it. The biomimicry of Echinocactus grusonii spines suggested an improved insertion strategy for self-inserting intraneural interfaces together with structural changes able to increase their elastic stability. The theoretical approach provided in this work was able to predict an increase of the first buckling threshold up to 39% for not puncturing self-inserting neural interfaces, and up to 59% for puncturing ones.

Skin meshing facilitates the greater expansion of donor skin through patterned slits and is widely used for treating extensive burn injuries. However, the actual expansion often falls below manufacturers' claims. Previous computational analyses using the isotropic Yeoh model have shown that Langer's line orientation and slit direction significantly affect induced stress and meshing ratios, yet the use of nonlinear anisotropic models that represent collagen fiber alignment corresponding to Langer's lines remains unexplored. This study employs a nonlinear anisotropic Gasser-Ogden-Holzapfel (GOH) model with slit orientations of 0°, 45°, and 90°, consistent with geometries reported in the literature, to quantify induced stress in skin meshing by incorporating collagen fibers within the dermis layer. The GOH parameters were calibrated to human back skin data uniaxially stretched parallel and perpendicular to Langer's lines using Levenberg-Marquardt optimization in the GIBBON toolbox (MATLAB R2023a) coupled with FEBio v4.0, achieving a standard deviation of 3% relative to experimental data. The GOH model predicted the highest induced stress at 100% strain for the 45° slit parallel to Langer's lines and the lowest for the 90° slit perpendicular, exceeding 40 MPa due to absence of damage and rupture modeling but accurately representing anisotropic mesh behavior.

Efficient coordination among software modules is essential for biomimetic robotic systems, much like the interaction among organs in a biological organism. However, implementing inter-process or inter-module communication in autonomous systems remains a complex and time-consuming task, particularly for new researchers. Simplifying inter-module communication is the central focus of this study. To address this challenge, we propose the DigitalTwinPort framework, a novel communication abstraction inspired by the port-based connectivity of embedded hardware systems. Unlike middleware-dependent solutions such as ROS, the proposed framework provides a lightweight, object-oriented structure that enables unified and scalable communication between software modules and networked devices. The concept is implemented in C++ and validated through an autonomous surface vehicle (ASV) developed for the RobotX Challenge. Results demonstrate that the DigitalTwinPort simplifies the development of distributed systems, reduces configuration overhead, and enhances synchronization between digital and physical components. This work lays the foundation for future digital twin architectures in embodied Internet systems, where software and hardware can interact seamlessly through standardized digital ports.