Flexible conductive materials possess excellent flexibility, stretchability, and electrical conductivity, enabling them to accommodate complex dynamic deformation modes, stretching, twisting, and wrinkling.This paper systematically reviews the structural characteristics, performance advantages, and fabrication features of the three major material systems, carbon-based materials, polymer composites, and novel flexible conductive materials, elucidates their electrical transport behavior and underlying mechanisms under mechanical deformation and cyclic strain, and compares the application limitations and shortcomings of different materials.Based on this, the paper summarizes current challenges regarding material performance synergy, long-term stability, processing costs, and large-scale fabrication, and analyzes future development directions such as composite structure design, mechanistic research, fabrication processes, and device integration. This paper systematically summarizes the common limitations across various device applications, providing a theoretical reference for the design and engineering applications of high-performance strain sensors and human-computer interaction devices.

This study strongly focuses on the topic "To what extent can mathematical modelling be used to calculate the maximum cornering speed of a Formula One car", which focuses on the practical application of mathematical modelling in the analysis of F1 car dynamics. Firstly, through the literature review, this paper introduced related concepts in physics and engineering. Next, modelling with different complexities is sorted out. Factors determining maximum cornering speed in F1 cars are also briefly introduced. Secondly, models are established, from the basic model of a mass point, the rigid body and the Paceijka Magic Formula. Thirdly, this paper analyzed them by theoretical derivation and chose independent situations to calculate theoretical results. Finally, it compared the percentage error with practical data and discussed the applicability and limitations, respectively. The results show that, with the increase of the complexity of the model, percentage error gradually decreases, much closer to the actual situation. However, at the same time, it also needs higher requirements in data analysis and handling. Therefore, mathematical modelling has strong explanatory ability here, but its scope of application still depends on the rationality of model assumptions. The conclusion of this study has a certain reference value for understanding the dynamics of racing cars and also helps to apply physics and math to practical engineering problems, which has a positive significance for learning vehicle dynamics and engineering modelling.

Soft robots represent an emerging class of robots characterized by strong adaptability, excellent flexibility, and high maneuverability. Today, they are widely used in fields such as bioengineering, disaster relief, industrial production, medical services, exploration, and surveying. Drive technology is a key factor limiting their performance. Based on a review and analysis of relevant domestic and international literature from recent years, this paper investigates three typical and common drive methods: pneumatic drive, hydraulic drive, and dielectric elastomer drive. The paper elucidates the principles and research progress of these three drive methods and compares their advantages and disadvantages from multiple perspectives, including response speed, execution accuracy, output power, and application scenarios. The results indicate that pneumatic drive technology is highly mature and offers balanced overall performance; hydraulic drive can achieve high output force and power but results in bulky equipment; and dielectric elastomer drive offers fast response speeds but requires a higher drive voltage. This study provides a theoretical reference for the selection of soft robot actuators and the research and development of hybrid drive systems.

Poleward mangrove planting is increasingly discussed as a coastal-restoration response to warming and habitat loss, yet planting at temperate estuarine sites may create trade-offs with existing tidal-flat and salt-marsh functions. This narrative review evaluates a prospective mangrove-planting scenario at Chongming Dongtan, with particular attention to crab-community reassembly and crab-mediated sediment and carbon processes. Evidence is separated into direct Chongming observations, adjacent evidence from eastern China, and mechanistic evidence from tropical and subtropical mangroves. Direct surveys show that the Dongtan crab assemblage is already dominated by the sesarmid Orisarma dehaani and the varunid Helice tientsinensis , while diking studies demonstrate that macrobenthic responses depend strongly on tidal connectivity. Mangrove planting could add structural and detrital resources, but dense or continuous stands could also contract open-flat habitat and redistribute rather than increase ecological functions. Three outcomes are proposed: habitat complementarity, functional substitution, and low-establishment engineering effects. Crabs may in turn alter planting trajectories through propagule consumption, bioturbation, and organic-matter redistribution. The review recommends replicated, reversible planting experiments that separate vegetation from engineering, retain mudflat and tidal-channel controls, and integrate crab traits, hydrology, sediment chemistry, greenhouse-gas fluxes, and remote sensing. The present evidence supports a testable adaptive experiment, not large-scale habitat conversion based on assumed benefits.

Benthic organisms have relatively limited mobility and long life cycles, enabling them to comprehensively reflect long-term changes in water quality, sediment conditions, and hydrological conditions. They are therefore important indicators for assessing the ecological health of lakes. This study integrates survey data on benthic animals in Dongting Lake from 2001 to 2025, with a particular focus on reviewing publicly available data from 2016 to 2025, and conducts stage-based visualization of three categories of indicators: phytoplankton, zooplankton, and benthic animals. The results show that the decline in total phosphorus concentrations and the improvement in water quality in Dongting Lake have provided a foundation for ecological restoration. However, publicly available data do not constitute a continuous time series collected annually using consistent methods. Data for the entire lake in 2020 and 2025 show that plankton density increased markedly, whereas benthic animal density remained relatively similar. Nevertheless, the Shannon-Wiener index of benthic animals was only 0.7 in 2025, suggesting that similar individual abundance does not necessarily indicate the recovery of community structure. Exploratory sampling in the nearshore waters of Junshan and Piaowei also revealed relatively few taxa and pronounced local dominance. Ecological restoration in Dongting Lake is jointly constrained by water quality, sediment conditions, hydrological connectivity, navigation, and shoreline habitats. Long-term monitoring based on fixed sampling sites, consistent seasons, and standardized taxonomic identification levels is therefore still needed.

Electric vertical take-off and landing (eVTOL) aircraft are central to the development of the low-altitude economy; however, the noise generated by their operation severely hampers the large-scale implementation of Urban Air Mobility (UAM) and significantly reduces public acceptance. This represents a critical bottleneck that urgently requires resolution for the commercialisation of the low-altitude transport industry. This paper focuses on the noise reduction technology framework for eVTOLs, systematically reviewing the current state of the art, engineering applications, existing challenges and future trends in this field. It provides an in-depth analysis of the noise reduction design and practical outcomes of the Joby Aviation S4, a representative model, thereby offering theoretical and practical support for the iterative development and engineering optimisation of eVTOL noise reduction technologies. The research employs a systematic literature review approach, combined with the analysis of representative case studies. The research findings indicate that the diverse range of existing noise-reduction technologies can effectively mitigate noise issues in eVTOL aircraft. The bespoke noise-reduction design of the Joby S4 has been validated through testing as delivering excellent noise-reduction performance, serving as a high-quality reference solution for the engineering implementation of noise reduction in civil eVTOL models. However, the industry currently faces numerous challenges, including insufficient accuracy in noise prediction, difficulties in balancing multiple performance metrics, significant challenges in the practical implementation of active noise cancellation technologies, and a lack of standardisation in technical standards.

This study compares Autoregressive Integrated Moving Average (ARIMA) and Long Short-Term Memory (LSTM) models for short-term forecasting of new confirmed COVID-19 cases in the United States. Using national-level data derived from the Johns Hopkins University CSSE COVID-19 repository, we evaluate both models under multiple experimental configurations that vary data frequency, training-test split, historical lookback length, in-put features, and model training strategy. Forecast performance is further compared with persistence and constant-growth benchmarks using rolling one-week-ahead evaluation. The results show that conclusions about model performance are highly sensitive to the experimental configuration. LSTM generally performed better than ARIMA in weekly forecasting, whereas ARIMA performed better under some daily-data settings. Additional epidemiological inputs and expanding-window retraining improved performance in several configurations. However, under the central weekly evaluation, both ARIMA and LSTM were outperformed by the simple persistence and constant-growth benchmarks. These findings demonstrate that increased model complexity does not necessarily lead to better short-term forecasting performance and highlight the importance of experimental design and strong baseline comparisons when evaluating time-series forecasting models.

Defect-parameter inversion requires numerical error, mode-localization constraints, and the conditioning of the inverse map to be considered separately. A two-dimensional square-lattice photonic crystal containing a point defect is studied using TM eigenmodes calculated with MIT Photonic Bands. Supercell-size and grid-resolution tests confirm frequency convergence, while the central energy fraction, inverse participation ratio, and effective mode area restrict the reliable calibration interval to 0.02 ≤ r_d/a ≤ 0.14, yielding 25 calibration samples. A PCHIP interpolant is compared with seven regression models. Nested leave-one-out cross-validation gives the lowest MAE for PCHIP (6.42 × 10⁻⁵ a), while BP-MLP performs best among the machine-learning models (1.47 × 10⁻⁴ a). Thirteen independent off-grid simulations confirm the accuracy of the leading nonlinear methods, although paired statistical tests do not distinguish BP-MLP, GPR, and SVR-RBF conclusively. Noise-propagation analysis further shows that practical inversion error is governed mainly by local inverse sensitivity, particularly near r_d/a = 0.02, where the forward frequency–radius slope is small. The resulting framework combines numerical convergence, physical mode screening, model comparison, and uncertainty analysis in a reproducible workflow for photonic-crystal parameter retrieval.

The generative AI has completely changed the creative industries by allowing users to create advanced visual content with minimal effort, which leads to many beginner artists skipping the basics, such as observance, proportion, perspective and correction. This article evaluates the limits of current generative AI in creative education and proposes an approach where generative AI becomes a diagnostic coach instead. Drawing from the idea of skill degradation due to automation, the paper shows how overdependence on image-generating technologies can undermine artists from developing their artistic expertise. It further discusses how imaging technologies can be applied to analyze the drawings and find errors in composition and structure with the necessary feedback provided in real time. The AI-based application for drawing education can help students identify drawing mechanics problems while making sure the learner has the ultimate responsibility in making artistic decisions. A human-based model also works in cases where AI directs attention to problems without taking their place. In this article, it is claimed that instead of positioning AI as a mere instrument of automatic output, it is more reasonable to consider it as a guide for effective learning, which guarantees that skills can be developed whilst preserving the autonomy of creativity. Nevertheless, there are difficulties related to bias in datasets, diversity of styles, and excessive standardization which should be taken into account to make sure that these applications do not inhibit the personal way of expressing oneself artistically.

Energy consumption, rather than path length, is the binding constraint for drones operating over dense urban terrain. This paper treats flight energy as the sum of three physically distinct terms: horizontal cruise, vertical climb, and attitude adjustment during turns. Three simplifying assumptions—constant cruise speed, negligible wind, and constant angular velocity in maneuvers—delimit where the model holds. On that foundation the paper develops an ant colony optimization variant whose heuristic scores the reciprocal of segment energy rather than distance, with Min-Max normalization across the factors and pheromone kept within bounds. Collision checking samples the segment interior at 5m intervals instead of testing endpoints only. For area coverage, the planning problem is split into a macro layer that sequences sub-regions and a micro layer that designs the scan pattern inside each one; transition costs between sub-regions are computed offline once and stored in a lookup table. Tests over Shenzhen Longhua District, partitioned into 30 sub-regions, cut total flight energy by 18.8 percent against baseline ACO while finishing 60.5 percent faster.