Aerospace Defense

Abstract In this study, we propose a novel multi-layer architecture, termed the Convolutional Sparse Network (CSN), designed to enhance fine-grained classification accuracy. This architecture is founded upon Dictionary Learning (DL) and Convolutional Sparse Coding (CSC). Fine-grained classification is considered one of the most challenging tasks in this domain due to significant intra-class variance and subtle inter-class similarities, necessitating the extraction of highly discriminative features. To achieve optimal performance, deep neural network-based models are typically engineered to localize informative regions at both the image and part levels, thereby extracting the discriminative features essential for precise classification. The proposed model maintains the capacity for high-level feature extraction, analogous to deep convolutional neural networks, while simultaneously mitigating information redundancy across the layer depth. These dual mechanisms facilitate a form of implicit localization, enabling the extraction of discriminative features that are highly effective for fine-grained tasks. Within this framework, we introduce a Sparse Autoencoder (SAE) which, when integrated with a standard pre-trained deep neural network, improves classification accuracy on fine-grained datasets. Experimental results demonstrate that the proposed hybrid model increases classification accuracy on the CUB-200 dataset by 3.5% compared to the base model and achieves a 28% reduction in the misclassification error rate.

Abstract Introduction: Mullite is a key high-temperature ceramic phase widely recognized for its outstanding properties in advanced refractory and structural applications.

Methods:This study investigates the synthesis and sintering behavior of mullite-based ceramics using mechanically activated Iranian andalusite. Raw andalusite powder was washed, characterized, and subjected to high-energy ball milling for up to ۶۰ h. The phase evolution, microstructural development, and sintering behavior of the milled powders were evaluated between ۸۰۰ °C and ۱۵۰۰ °C and compared with an unmilled reference sample.

Findings: XRD analysis showed that prolonged mechanical activation disrupted the crystalline structure of andalusite and lowered the onset of mullitization by approximately ۴۰۰ °C. The activated powder began forming mullite at ۸۰۰ °C and reached a predominantly single-phase mullite state above ۱۲۰۰ °C. Crystallite size calculated by the Scherrer equation remained in the nanocrystalline range (approximately ۸–۱۳ nm), while Williamson–Hall analysis indicated a progressive reduction of lattice microstrain with increasing temperature, approaching near-zero values at ۱۵۰۰ °C. SEM observations of samples sintered at ۱۵۰۰ °C revealed that mechanical activation produced finer, more elongated, and more uniformly distributed acicular mullite grains (aspect ratio ≈ ۸–۱۵) together with a denser microstructure, in contrast to the coarser and more porous structure of the unmilled sample.

Conclusion:Consequently, the activated powder exhibited more stable densification and higher relative density at lower firing temperatures. This work provides an efficient route for fabricating fine-grained, homogeneous mullite ceramics that are of significant interest for high-temperature aerospace and defense applications

Abstract Introduction:The development of advanced functional nanomaterials is crucial for next-generation optoelectronic devices. Germanium Carbide Nanotubes (GeCNTs) have emerged as promising candidates due to their unique electronic properties and potential for structural tunability, making them highly relevant for high-tech applications. Methods:In this study, the electronic and optical properties of GeCNTs with various chiral indices were investigated under the influence of hydrostatic strain. The calculations were performed using first-principles methods based on Density Functional Theory (DFT) within the Perdew-Burke-Ernzerhof Generalized Gradient Approximation (PBE-GGA) framework. Findings:The results demonstrate a significant dependence of electronic behavior on the nanotube chirality. Specifically, the (5,0) GeCNT exhibits a near-zero direct bandgap, behaving as a quasi-metallic structure, whereas the (10,0) GeCNT functions as a semiconductor with a direct bandgap of 1.29 eV. Optical property analysis in out-of-plane polarization reveals that the most significant optical transitions occur within the 1.5 to 4 eV energy range. Furthermore, both changes in chirality and the application of hydrostatic strain lead to substantial modifications in the optical response of these nanostructures. Conclusion:GeCNTs offer exceptional tunability in both electronic and optical domains. These characteristics position GeCNTs as highly suitable materials for applications in optical sensors, photonic devices, and infrared detectors, particularly within the aerospace and defense sectors.

Abstract Introduction: Quadrotor unmanned aerial vehicles (UAVs) are highly nonlinear, underactuated, and strongly coupled systems whose performance is sensitive to physical parameter uncertainties. Among these uncertainties, mass variation is especially important because it directly affects the altitude dynamics, since the vertical acceleration depends on the ratio between the applied thrust and the vehicle mass. In practical missions, the quadrotor mass may change due to payload variation, battery replacement, mounted equipment, or modeling errors. Therefore, designing a stabilizing controller that does not require exact prior knowledge of the vehicle mass is essential for improving flight reliability and robustness.
Methods: In this study, a nonlinear mathematical model of the quadrotor is first derived using the Newton–Euler formulation. The model includes translational and rotational dynamics and is represented in a twelve-state nonlinear state-space form. Then, a nominal integrator backstepping controller is designed for altitude and attitude stabilization under the assumption that all physical parameters are known. To handle unknown mass uncertainty, the nominal controller is extended into an adaptive integrator backstepping controller.
Findings: Numerical simulations show that the proposed controller drives the quadrotor altitude and the roll, pitch, and yaw angles to their reference values with zero steady-state error and no overshoot in the attitude channels. Benchmarking against non-adaptive integrator backstepping, PID, MPC, and SMC under identical mass uncertainty shows that the proposed scheme, unlike its non-adaptive counterpart, eliminates the steady-state altitude error (final altitude of 0.998 m versus 0.626 m) and is the only controller that reconstructs the unknown and even time-varying mass online; under a +50% mass variation it is the only method that fully reaches the altitude reference (ITAE of 3.85 versus 17.43 for SMC).
Conclusion: The proposed robust adaptive integrator backstepping controller provides an effective solution for quadrotor stabilization in the presence of unknown mass uncertainty. By combining Lyapunov-based adaptive estimation with integrator backstepping control, the method guarantees boundedness of closed-loop signals and improves robustness against mass variations, making it suitable for quadrotor applications involving payload changes or uncertain modeling conditions.

Abstract Introduction: Shear buckling is considered one of the primary instability modes in sandwich panels with flexible cores. The negligible shear stiffness of the core significantly reduces the panel’s stability against shear loads, while the occurrence of thickness-direction nonlinear deformations and core compressibility complicates the buckling mechanism, thereby challenging the simplifying assumptions of conventional theories.

Methods: In this study, an improved higher-order sandwich panel theory is developed for the stability analysis of thick panels featuring a flexible core, wherein the displacement fields of the face sheets and the core are defined independently as higher-order functions along the thickness direction. Furthermore, the stress compatibility conditions and displacement continuity at the face sheet–core interfaces are rigorously enforced, thereby capturing the intricate bending–shear–extension couplings with a constant number of unknown kinematic variables. The governing equations are derived via the principle of minimum potential energy and subsequently solved using the Galerkin method.

Findings: Numerical results revealed that the proposed theory estimates the shear buckling load lower than the HSAPT by 2.3% for symmetric square panels, 1.9% for unsymmetric square panels, 2.1% for symmetric rectangular panels, and 2.0% for unsymmetric rectangular panels. Furthermore, this theory exhibits an enhanced capability to distinguish and predict localized wrinkling modes. The findings also indicated that face sheet thickness asymmetry reduces the critical buckling load due to the degradation of effective panel stiffness.

Conclusion: The reduced critical loads yielded by the present formulation are attributed to the rigorous satisfaction of inter-component continuity conditions alongside core shear compliance, thereby correcting the unconservative estimations inherent in prior models. Hence, the proposed model provides a reliable, high-precision framework for the safe design and structural optimization of sandwich panels under in-plane shear conditions.