Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 6
Encontrada(s) 1 página(s)

Investigo. Repositorio Institucional de la Universidade de Vigo
Publicación Artículo científico (JournalArticle)

Projection of compound wind and precipitation extreme events in the Iberian Peninsula based on CMIP6

  • Álvarez Fernández, María Inés
  • Pereira, H.
  • Picado, A.
  • Sousa, M. C.
  • Lorenzo González, María de las Nieves
  • Dias, J. M.
This study aims to investigate the potential changes in the co-occurrence of strong precipitation and wind events over the Iberian Peninsula using simulations from the Coupled Model Intercomparison Project (CMIP) Phase 6 under two scenarios (SSP2 − 4.5 and SSP5 − 8.5). Projected changes indicate a significant regional variability during all seasons. In winter, the western regions are projected to experience an increase in compound events as the century progresses under both scenarios, with a significantly larger area being affected by the end of the century. In spring, summer, and autumn, a general decline in the occurrence of these events is anticipated throughout the century, accompanied by a reduction in the area affected by them. However, in the northwesternmost area (Galicia), an increase in the occurrence of compound events is expected during the spring towards the end of the century, particularly under the SSP5-8.5 scenario., Xunta de Galicia | Ref. ED431C 2021/44, Agencia Estatal de Investigación | Ref. TED2021-129152B-C43, Fundação para a Ciência e a Tecnologia, Universidade de Vigo/CISUG
Investigo. Repositorio Institucional de la Universidade de Vigo
Publicación Artículo científico (JournalArticle)

A digital-twin LiDAR simulator for performance assessment of railway ballast geometry inspections

  • Ríos Otero, Eduardo
  • Aldao Pensado, Enrique
  • Fernández-Pardo, L.M.
  • Fontenla Carrera, Gabriel
  • Veiga López, Fernando
  • González Jorge, Higinio
High-speed railways are increasingly preferred for medium-distance travel, thanks to their efficiency and convenience. However, the growing frequency of services accelerates infrastructure wear, raising inspection and maintenance demands. To address this, autonomous inspection solutions based on LiDAR (Light Detection and Ranging) systems and onboard cameras have recently been proposed, but their testing and deployment remain challenging due to regulatory constraints and the difficulty of accessing representative railway environments. This work presents a LiDAR digital twin that realistically replicates sensor behaviour in real railway track environments. The simulator incorporates CAD (Computer-Aided Design) models of the track vehicle and sensor installation, enabling the assessment of installation-related limitations and sensor field-of-view occlusions. Two commercial sensors, the Livox Avia and Livox HAP, were metrologically calibrated using experimental data acquired on a real railway track. Good agreement between simulated and experimental data was observed, with deviations of only 2 points/cm in point density and centimetre-level differences in reconstructed ballast geometry. The digital twin was subsequently used to assess the suitability of this technology for ballast geometry measurement at different travel speeds, showing centimetre-order errors for both sensors at speeds of up to 120 km/h., Universidade de Vigo/CISUG, Agencia Estatal de Investigación | Ref. PID2024-156356OB-I00, Agencia Estatal de Investigación | Ref. PID2021-125060OB-I00, Xunta de Galicia | Ref. GPC-ED431B 2025/31, Agencia Estatal de Investigación | Ref. CPP2022-009679
Investigo. Repositorio Institucional de la Universidade de Vigo
Publicación Artículo científico (JournalArticle)

Heat index historical trends and projections due to climate change in the Mediterranean basin based on CMIP6

  • Álvarez Fernández, María Inés
  • Diaz Poso, A.
  • Lorenzo González, María de las Nieves
  • Roye, D.
Air temperature and relative humidity can be considered as two essential meteorological parameters in the determination of heat stress. The heat index (HI) includes both of them and it is appropriate for determining the thermal conditions of different climates. We investigated potential changes in the HI for the Mediterranean basin using simulations from the Coupled Model Intercomparison Project Phase 6 (CMIP6) climate models under two future scenarios (Shared Socio-economic Pathways: SSP2-4.5 and SSP5-8.5) over the period 2020–2099. Results reveal an important increase of HI at the end of the 21st century for both scenarios, with greater changes for the SSP5-8.5 scenario all over the basin. Strong significant upwards trends (around 1 °C per decade; significance level computed at 5%) are expected in the entire area and for all months at the end of the century, with greatest values during the summer months (close to 1.5 °C per decade) along the coastal areas of the basin. Many areas of the Southern Mediterranean basin (Africa and Arabian Peninsula) will be strongly affected with dangerously high heat index values (higher than 41 °C) during summer months by the end of the 21st century. A northward extension of these dangerous conditions is also expected including several areas of southern Europe, Xunta de Galicia | Ref. ED431C 2021/44, Agencia Estatal de Investigación | Ref. PID2021-125060OB-I00, Agencia Estatal de Investigación | Ref. TED2021-129152B-C43, Agencia Estatal de Investigación | Ref. TED2021-129756B-C31, Universidade de Vigo/CISUG, European Union | Ref. PRTR-C17.I1
Investigo. Repositorio Institucional de la Universidade de Vigo
Publicación Artículo científico (JournalArticle)

Learning heuristics with Vision Transformers for risk-aware drone last-mile delivery

  • Aldao Pensado, Enrique
  • Veiga López, Fernando
  • Chanel, Caroline P.C.
  • Watanabe, Yoko
  • González Jorge, Higinio
for urban parcel transport, offering a sustainable and cost-effective alternative to traditional ground-based logistics. Such operations, however, introduce significant risks, requiring effective mitigation strategies and efficient path-planning tools. Risk-aware path planning in complex urban environments remains computationally demanding, as it involves a highly combinatorial sequential decision-making problem while requiring efficient online replanning to ensure reactivity. Efficient planners often rely on graph-based search methods, in which the heuristic used to guide the search plays a crucial role. However, in risk-aware path planning, heuristic estimation is particularly challenging, as risk evaluation forms part of the cost estimation and differs significantly from classical line-of-sight or Euclidean approaches. To address this limitation, this study presents a novel Deep Learning framework for estimating accurate and admissible heuristic functions in risk-aware UAV path planning. To this end, a comprehensive database of cost-to-go maps was generated using realistic 3D urban scenarios and a graph-based optimisation framework. This dataset was used to train a Vision Transformer (ViT) network capable of providing accurate cost-to-go estimates for various objective-function configurations. The proposed architecture incorporates an admissibility correction mechanism that regulates the trade-off between heuristic accuracy and admissibility, promoting either solution optimality or search efficiency. The method was benchmarked against conventional heuristics in terms of computation time and cost-to-go. Results show speed-ups of up to 20 times compared to traditional approaches while maintaining near-optimal solutions, with marginal increases in average path cost of the order of 0.1%., Agencia Estatal de Investigación | Ref. PID2024-156356OB-I00, Agencia Estatal de Investigación | Ref. CPP2022-009679, Ministerio de Ciencia, Innovación y Universidades | FPU21/ 01176, Xunta de Galicia | Ref. FEADER2024/052B, Xunta de Galicia | Ref. ED431B 2025/31, Ministerio de Defensa | Ref. 2019 65 0090004707501, Universidade de Vigo/CISUG
Investigo. Repositorio Institucional de la Universidade de Vigo
Publicación Artículo científico (JournalArticle)

Synthetic data generation techniques for enhancing crack detection in railway concrete sleepers

  • Aldao Pensado, Enrique
  • Fernández Pardo, L.
  • Veiga López, Fernando
  • González de Santos, L. M.
  • González Jorge, Higinio
Developing deep learning–based computer vision systems for the autonomous inspection of railway sleepers is a challenging task due to the difficulty in obtaining training images representative of the actual infrastructure. To address this issue, this work proposes two different synthetic data generation methodologies. The first method involves a neural network based on Stable Diffusion 1.5, a type of latent diffusion model (LDM) capable of generating photorealistic images through a process that sequentially applies denoising autoencoders. Fine-tuning of this model was performed using images of real railway tracks to generate pictures of damaged concrete sleepers. The second proposed method is a Crack Simulator based on image processing techniques. It modifies images of undamaged sleepers by introducing textures and morphologies of cracks, which are randomly generated using one-dimensional Perlin noise signals. Additionally, it incorporates semantic classification maps of track elements to determine the origin and propagation zones of the cracks, considering factors such as ballast stone occlusions. The main objective of this study is to evaluate the capabilities of these synthetic approaches and compare their effectiveness with respect to the use of generic open-source databases. To this end, synthetic image data sets were created, and several YOLO (you only look once) object detection models were trained. The performance of the detectors was evaluated, and a quantitative assessment of each synthetically generated image data set was conducted. Results showed that the Stable Diffusion model exhibits remarkable potential in improving the concrete damage detection performance of deep learning systems., Ministerio de Universidades | Ref. FPU21/01176, Agencia Estatal de Investigación | Ref. PLEC2021-007940, Agencia Estatal de Investigación | Ref. CPP2022-009679
Investigo. Repositorio Institucional de la Universidade de Vigo
Publicación Artículo científico (JournalArticle)

Assessing the performance of the CERRA dataset in reproducing precipitation and extreme precipitation events over the Iberian Peninsula

  • Álvarez Fernández, María Inés
  • Pereira, H.
  • Lorenzo González, María de las Nieves
  • Picado, A.
  • Dias, J. M.
Study region The Iberian Peninsula (IP). Study focus Understanding precipitation and extreme precipitation events is crucial for managing climate impacts on human activities, ecosystems, and environment. Reanalysis datasets such as ERA-Interim and ERA5 are commonly used to studying these patterns; however, their spatial resolution often limits their accuracy. The recent high-resolution Copernicus European Regional ReAnalysis (CERRA) product provides an opportunity to improve precipitation estimates, particularly in regions with complex topography. The present study aims to assess the performance of the CERRA dataset in representing precipitation and extreme precipitation events across the IP. The analysis involves a comparison between CERRA and observational datasets, focusing on several annual and seasonal spatial precipitation indices. Special attention is given to the representation of extreme precipitation and their associated trends. New hydrological insights for the region The results indicate that CERRA effectively captures the spatial distribution of precipitation across most seasons, with weaker performance during the summer. CERRA reliable represents extreme precipitation and their temporal trends across the IP, especially in regions with complex orography such as Galicia and the Pyrenees. However, it tends to overestimate precipitation in drier areas. Trend analysis also revealed that, while CERRA successfully reflects the general patterns of precipitation extremes, some discrepancies remain, especially in autumn. Overall, CERRA demonstrates strong potential as a valuable dataset for hydrological and climatological studies over the IP., Fundação para a Ciência e a Tecnologia | Ref. UIDP/50017/2020, Fundação para a Ciência e a Tecnologia | Ref. UIDB/50017/2020, Fundação para a Ciência e a Tecnologia | Ref. LA/P/0094/2020, Xunta de Galicia | Ref. ED431C 2021/44, Agencia Estatal de Investigación | Ref. PID2021–125060OB-I00, Agencia Estatal de Investigación | Ref. TED2021–129756B-C31