Roberto Gómez-Espinosa Martin

Mr. Roberto Gómez-Espinosa Martin

Artificial Intelligence, HI Iberia (Spain), Madrid, Spain
orcid

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
7 February 2023 in Drones
Verified
Journal Article
Open access
PDF file
MNManuel Naranjo
Manuel Naranjo
DFDiego Fuentes
Diego Fuentes
ECElena Muelas Cano
Elena Muelas Cano
EBEnrique Díez Benito
Enrique Díez Benito
LCLuis Ciruelo
Luis Ciruelo
ILInmaculada Luengo
Inmaculada Luengo
The construction industry is on the path to digital transformation. One of the main challenges in this process is inspecting, assessing, and maintaining civil infrastructures and construction elements. However, Artificial Intelligence (AI) and Unmanned Aerial Vehicles (UAVs) can support the tedious and time-consuming work inspection processes. This article presents an innovative object detection-based system which enables the detection and geo-referencing of different traffic signs from RGB images captured by a drone’s onboard camera, thus improving the realization of road element inventories in civil infrastructures. The computer vision component follows the typical methodology for a deep-learning-based SW: dataset creation, election and training of the most accurate object detection model, and testing. The result is the creation of a new dataset with a wider variety of traffic signs and an object detection-based system using Faster R-CNN to enable the detection and geo-location of traffic signs from drone-captured images. Despite some significant challenges, such as the lack of drone-captured images with labeled traffic signs and the imbalance in the number of images for traffic signal detection, the computer vision component allows for the accurate detection of traffic signs from UAV images.

25 February 2025 in Particles
Verified
Journal Article
Open access
PDF file
GRGalo Gallardo Romero
Galo Gallardo Romero
GLGuillermo Rodríguez Llorente
Guillermo Rodríguez Llorente
LRLucas Magariños Rodríguez
Lucas Magariños Rodríguez
RNRodrigo Morant Navascués
Rodrigo Morant Navascués
NPNikita Khvatkin Petrovsky
Nikita Khvatkin Petrovsky
Roberto Gómez-Espinosa MartinRoberto Gómez-Espinosa Martin
Roberto Gómez-Espinosa Martin
One of the primary challenges for future nuclear fusion power plants is understanding how neutron irradiation affects reactor materials. To tackle this issue, the IFMIF-DONES project aims to build a facility capable of generating a neutron source in order to irradiate different material samples. This will be achieved by colliding a deuteron beam with a lithium jet. In this work, within the DONES-FLUX project, deep learning surrogate models are applied to the design and optimization of the IFMIF-DONES linear accelerator. Specifically, neural operators are employed to predict deuteron beam envelopes along the longitudinal axis of the accelerator and neutron irradiation effects at the end, after the beam collision. This approach has resulted in models that are able of approximating complex simulations with high accuracy (less than 17% percentage error for the worst case) and significantly reduced inference time (ranging from 2 to 6 orders of magnitude) while being differentiable. The substantial speed-up factors enable the application of online reinforcement learning algorithms, and the differentiable nature of the models allows for seamless integration with differentiable programming techniques, facilitating the solving of inverse problems to find the optimal parameters for a given objective. Overall, these results demonstrate the synergy between deep learning models and differentiable programming, offering a promising collaboration among physicists and computer scientists to further improve the design and optimization of IFMIF-DONES and other accelerator facilities. This research will lay the foundations for future projects, where optimization efforts with differentiable programming will be performed.

7 February 2023 in arXiv
Verified
Preprint
Open access
PDF file
EAEduardo Abenza
Eduardo Abenza
CACésar Alonso
César Alonso
ISIsabel Sobrados
Isabel Sobrados
JAJosé M. Amarilla
José M. Amarilla
JRJavier L. Rodríguez
Javier L. Rodríguez
MAMaria-Carmen Asensio
Maria-Carmen Asensio
A significant step forward in Lithium-ion batteries (LIBs) developments can only be achieved by proposing mold-breaking research based on selecting the best materials for the cell components, optimizing cell manufacture, anticipating the degradation mechanisms of the LIBs, and consolidating the regeneration processes of damaged batteries. LIBs with longer recycling life, better safety, and the ability to be reused will establish sustainable state-of-the-art batteries with maximum energy efficiency, low costs, and minimal CO2 emissions within a circular economy, promoting sustainability in areas as relevant as electromobility and portable electronics. Recently, there has been increasing interest in applying Artificial Intelligence (AI) techniques and their subclasses to better predict novel materials with designed properties. This collection of methods has already obtained considerable success, having been used to predict numerous physical properties of materials. However, compared to other fields, the materials data are typically much smaller and sometimes more diverse, which undoubtedly affects the construction and effectiveness of AI models. At present, several Data Augmentation (DA) methods have been proposed in materials science based on flipping, rotating, and distorting the unit cells of materials, which have been demonstrated to be very efficient in increasing the size and quality of data. Here we present an even more effective new method of Data Augmentation based on the lattice scaling of crystal structures. In the lattice scaling DA method, the unit cell is perturbed, undergoing an increase and decrease of the unit cell volume in an isotropic or anisotropic way. This transformation is particularly pertinent for battery components since volume changes of up to 5% have been reported for the insertion-based LIBs during cycling.

1 May 2026 in Journal of Physics: Conference Series
Verified
Conference Paper
Open access
GLGuillermo Rodríguez Llorente
Guillermo Rodríguez Llorente
JGJose María Moreu Gamazo
Jose María Moreu Gamazo
GNGuillermo Silvente Niñirola
Guillermo Silvente Niñirola
IBIrene Barrera Bernand
Irene Barrera Bernand
Roberto Gómez-Espinosa MartinRoberto Gómez-Espinosa Martin
Roberto Gómez-Espinosa Martin
Floating wind structures are exposed to highly dynamic metocean conditions that accelerate fatigue, raise failure risks, and increase operational costs. Current predictive maintenance strategies rely heavily on computationally expensive simulations, which are unsuitable for real-time operation at farm scale. The e-PROA project addresses this challenge by developing a digital twin powered by surrogate models, enabling an accurate and fast prediction of structural responses. Fourier Neural Operator (FNO) surrogate models were trained on datasets generated with OrcaFlex and AQWA under diverse sea states. These models replace time-intensive simulations with efficient approximations, allowing prediction of mooring line tensions and tower bending moments in near real-time. A comprehensive sensor configuration study involving 255 FNO models per use case determined optimal sensor placement, achieving a MAPE of 0.333% for axial forces and 3.122% for bending moments. Multi-condition models validated against unseen sea states confirmed generalization ability with MAPE consistently below 5%. Compared to full physics-based simulations, surrogate models reduce computational time by several orders of magnitude, making real-time deployment feasible. The system is designed to incorporate anomaly detection, remaining useful life (RUL) estimation, and decision-support dashboards, providing operators with clear and interpretable information for proactive maintenance of floating offshore wind farms.

29 September 2026 in EPJ Research Infrastructures
Verified
Journal Article
Open access
PDF file
GLGuillermo Rodríguez Llorente
Guillermo Rodríguez Llorente
GRGalo Gallardo Romero
Galo Gallardo Romero
RNRodrigo Morant Navascués
Rodrigo Morant Navascués
ASAnderson Sabogal
Anderson Sabogal
CTClaudio Torregrosa-Martín
Claudio Torregrosa-Martín
Roberto Gómez-Espinosa MartinRoberto Gómez-Espinosa Martin
Roberto Gómez-Espinosa Martin
Developing materials capable of withstanding extreme irradiation is a fundamental challenge for realizing nuclear fusion as a sustainable energy source. The IFMIF–DONES facility, currently under construction, will address this need by producing a fusion-relevant neutron flux using a high-power deuteron beam striking a flowing lithium target, enabling accelerated qualification of candidate structural materials. Validating the accelerator and target technologies requires precise control of the ultra-high-vacuum region near the lithium surface, yet direct pressure sensing in this irradiated zone is challenging and conventional controllers can struggle with the associated nonlinear gas-flow dynamics. As part of the DONES–FLUX project, this work introduces a data-driven workflow for indirect pressure estimation and autonomous regulation in the MuVacAS prototype, a dedicated experimental platform that reproduces the final beamline section of IFMIF–DONES. Two Fourier Neural Operator surrogate models are learned from real data: one serves as a virtual sensor to infer chamber pressure from upstream readings, while the other models system dynamics to provide a safe, fast training environment for a Deep Reinforcement Learning controller. The trained agent uses virtual-sensor feedback during deployment and reliably regulates argon injection across operating scenarios, including conditions not seen during training. Experimental campaigns show that the RL-based controller yields promising control results, achieving stable regulation without manual retuning. These results demonstrate a viable pathway for deploying advanced learning-based control in fusion-relevant accelerator systems, supporting the operational readiness of IFMIF–DONES and future fusion technology infrastructures.

17 December 2025 in arXiv
Verified
Preprint
Open access
PDF file
GLGuillermo Rodríguez Llorente
Guillermo Rodríguez Llorente
GRGalo Gallardo Romero
Galo Gallardo Romero
RNRodrigo Morant Navascués
Rodrigo Morant Navascués
NPNikita Khvatkin Petrovsky
Nikita Khvatkin Petrovsky
ASAnderson Sabogal
Anderson Sabogal
Roberto Gómez-Espinosa MartinRoberto Gómez-Espinosa Martin
Roberto Gómez-Espinosa Martin
The development of nuclear fusion requires materials that can withstand extreme conditions. The IFMIF-DONES facility, a high-power particle accelerator, is being designed to qualify these materials. A critical testbed for its development is the MuVacAS prototype, which replicates the final segment of the accelerator beamline. Precise regulation of argon gas pressure within its ultra-high vacuum chamber is vital for this task. This work presents a fully data-driven approach for autonomous pressure control. A Deep Learning Surrogate Model, trained on real operational data, emulates the dynamics of the argon injection system. This high-fidelity digital twin then serves as a fast-simulation environment to train a Deep Reinforcement Learning agent. The results demonstrate that the agent successfully learns a control policy that maintains gas pressure within strict operational limits despite dynamic disturbances. This approach marks a significant step toward the intelligent, autonomous control systems required for the demanding next-generation particle accelerator facilities.