Pasar al contenido principal

Resultados relacionados

Resultados totales (Incluyendo duplicados): 9
Encontrada(s) 1 página(s)

e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Electric Vehicle Route Optimization: An End-to-End Learning Approach with Multi-Objective Planning

  • Gutiérrez Moreno, Rodrigo
  • Llamazares Llamazares, Ángel|||0000-0001-8273-5163
  • Revenga de Toro, Pedro Alfonso|||0000-0003-2550-5972
  • Ocaña Miguel, Manuel|||0000-0002-8875-1866
  • Antunes García, Miguel|||0009-0008-5627-5325
Traditional routing algorithms optimizing for distance or travel time are inadequate for electric vehicles (EVs), which require energy-aware planning considering battery constraints and charging infrastructure. This work presents an energy-optimal routing system for EVs that integrates personalized consumption modeling with real-time environmental data. The system employs a Long Short-Term Memory (LSTM) neural network to predict State-of-Charge (SoC) consumption from real-world driving data, learning directly from spatiotemporal features including velocity, temperature, road inclination, and traveled distance. Unlike physics-based models requiring difficult-to-obtain parameters, this approach captures nonlinear dependencies and temporal patterns in energy consumption. The routing framework integrates static map data, dynamic traffic conditions, weather information, and charging station locations into a weighted graph representation. Edge costs reflect predicted SoC drops, while node penalties account for traffic congestion and charging opportunities. An enhanced A* algorithm finds optimal routes minimizing energy consumption. Experimental validation on a Nissan Leaf shows that the proposed end-to-end SoC estimator significantly outperforms traditional approaches. The model achieves an RMSE of 36.83 and an ?2 of 0.9374, corresponding to a 59.91% reduction in error compared to physics-based formulas. Real-world testing on various routes further confirms its accuracy, with a Mean Absolute Error in the total route SoC estimation of 2%, improving upon the 3.5% observed for commercial solutions., European Union, Ministerio de Ciencia, Innovación y Universidades, Agencia Estatal de Investigación, Comunidad de Madrid
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Point Cloud Painting for 3D Object Detection with Camera and Automotive 3+1D RADAR Fusion

  • Montiel Marín, Santiago|||0009-0000-5492-0839
  • Llamazares Llamazares, Ángel|||0000-0001-8273-5163
  • Antunes García, Miguel|||0009-0008-5627-5325
  • Revenga de Toro, Pedro Alfonso|||0000-0003-2550-5972
  • Bergasa Pascual, Luis Miguel|||0000-0002-0087-3077
RADARs and cameras have been present in automotives since the advent of ADAS, as they possess complementary strengths and weaknesses but have been underlooked in the context of learning-based methods. In this work, we propose a method to perform object detection in autonomous driving based on a geometrical and sequential sensor fusion of 3+1D RADAR and semantics extracted from camera data through point cloud painting from the perspective view. To achieve this objective, we adapt PointPainting from the LiDAR and camera domains to the sensors mentioned above. We first apply YOLOv8-seg to obtain instance segmentation masks and project their results to the point cloud. As a refinement stage, we design a set of heuristic rules to minimize the propagation of errors from the segmentation to the detection stage. Our pipeline concludes by applying PointPillars as an object detection network to the painted RADAR point cloud. We validate our approach in the novel View of Delft dataset, which includes 3+1D RADAR data sequences in urban environments. Experimental results show that this fusion is also suitable for RADAR and cameras as we obtain a significant improvement over the RADAR-only baseline, increasing mAP from 41.18 to 52.67 (+27.9%)., Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

A game-theoretic approach to mitigate charging anxiety for electric vehicle users through multi-parameter dynamic pricing and real-time traffic flow

  • Rasheed, Muhammad Babar|||0000-0002-9911-0693
  • Llamazares Llamazares, Ángel|||0000-0001-8273-5163
  • Ocaña Miguel, Manuel|||0000-0002-8875-1866
  • Revenga de Toro, Pedro Alfonso|||0000-0003-2550-5972
While price-based demand response is universally recognized as crucial for managing electric vehicle charging load, the academic literature has explored diverse mechanisms for its implementation. Prior research has revealed that applying load management schemes based on price-based demand response programs results in higher scheduling costs for low or constant-energy consumers. In addition, the inherent uncertainties involved in arrival, service requirements, market pricing and user behaviours have posed serious challenges in managing real-time charging operations. To handle these challenges, this work has addressed the charging problem at the scheduling level by analysing arrival, waiting, service and departure requests. This work proposed a mathematical model to coordinate charging to minimize the total charging cost and time for a given number of vehicles. We have developed an advanced charging algorithm and introduced a multi-leader multi-follower Stackelberg game. This strategic model ensures a Nash equilibrium, aligning the interests of consumers (vehicles) and providers (charging stations), ensuring a fair and balanced charging system. Where, charging load demand, real-time pricing, charging location, and vehicle types are used as input parameters. In the first stage, the real-time pricing obtained from the Spanish Electricity network is used to schedule the charging load, whereas the second stage is dedicated to using the proposed multi-parameter pricing for customer satisfaction and reduced cost. The results demonstrate the superior performance of our proposed scheme compared to existing approaches. Finally, we present a real-life case study to illustrate the practical effectiveness of the proposed techniques. The study underscores the potential of our approach in real-world scenarios, offering a promising avenue for enhancing efficiency and user experience in charging systems., Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

SalsaNext+: a multimodal-based point cloud semantic segmentation with range and RGB images

  • Sánchez García, Fabio
  • Montiel Marín, Santiago|||0009-0000-5492-0839
  • Antunes García, Miguel|||0009-0008-5627-5325
  • Gutiérrez Moreno, Rodrigo
  • Llamazares Llamazares, Ángel|||0000-0001-8273-5163
  • Bergasa Pascual, Luis Miguel|||0000-0002-0087-3077
Advances in sensor fusion techniques are redefining the landscape of 3D point cloud semantic segmentation, particularly for autonomous driving applications. We propose an enhanced approach that leverages the complementary strengths of LiDAR and multi-camera systems. This study introduces two extensions to the state-of-the-art SalsaNext model based only in LiDAR: SalsaNext+RGB, which integrates RGB data into range-view (RV) images, and SalsaNext+PANO, incorporating panoramic images built from multi-camera setups. The proposed methods are evaluated using the SemanticKITTI and Panoptic nuScenes datasets, showing notable improvements in segmentation accuracy. Results indicate that RGB fusion boosts performance with minimal latency, while panoramic integration offers additional gains at the expense of higher computational load. Comparative analyses highlight significant mIoU gains, demonstrating the potential of multimodal sensor fusion for intricate driving scene understanding., Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Contribución de congreso (ConferenceOutput)

Fast and Efficient Transformer-based Method for Bird's Eye View Instance Prediction

  • Antunes García, Miguel|||0009-0008-5627-5325
  • Bergasa Pascual, Luis Miguel|||0000-0002-0087-3077
  • Montiel Marín, Santiago|||0009-0000-5492-0839
  • Barea Navarro, Rafael|||0000-0002-4179-6100
  • Sánchez García, Fabio
  • Llamazares Llamazares, Ángel|||0000-0001-8273-5163
2024 IEEE 27th International Conference on Intelligent Transportation Systems (ITSC), 24-27 September 2024, Edmonton, AB, Canada., Accurate object detection and prediction are critical to ensure the safety and efficiency of self-driving architectures. Predicting object trajectories and occupancy enables autonomous vehicles to anticipate movements and make decisions with future information, increasing their adaptability and reducing the risk of accidents. Current State-Of-The-Art (SOTA) approaches often isolate the detection, tracking, and prediction stages, which can lead to significant prediction errors due to accumulated inaccuracies between stages. Recent advances have improved the feature representation of multi-camera perception systems through Bird'Eye View (BEV) transformations, boosting the development of end-to-end systems capable of predicting environmental elements directly from vehicle sensor data. These systems, however, often suffer from high processing times and number of parameters, creating challenges for real-world deployment. To address these issues, this paper introduces a novel BEV instance prediction architecture based on a simplified paradigm that relies only on instance segmentation and flow prediction. The proposed system prioritizes speed, aiming at reduced parameter counts and inference times compared to existing SOTA architectures, thanks to the incorporation of an efficient transformer-based architecture. Furthermore, the implementation of the proposed architecture is optimized for performance improvements in PyTorch version 2.1., Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Contribución de congreso (ConferenceOutput)

Design and development of a digital twin for monitoring railway infrastructure

  • Fuentes Puche, Javier
  • Fierro Chicaiza, Franck Michael
  • Barea Navarro, Rafael|||0000-0002-4179-6100
  • López Guillén, María Elena
  • Bergasa Pascual, Luis Miguel|||0000-0002-0087-3077
2025 IEEE Intelligent Vehicles Symposium (IV), 22-25 June 2025, Cluj-Napoca, Romania., 6 p., The aim of this work is to develop a digital twin application to ensure an optimal level of reliability when launching a larger project based on the identification of trains and detection of defects that allow for safe freight transport on spanish trains. The digital twin (DT) framework consists of three parts: the ?physical product? which consists of a scanning camera placed on a track gantry, the ?virtual product? which includes a model based on real-time data representing the freight car detected by the perception system, and the data flow connections. The camera images will be post-processed through an artificial intelligence detection model (YOLOv8), trained to detect all the elements necessary for the safety of the vehicle and the cargo. Field studies have demonstrated the effectiveness of the proposed digital twin framework and its potential to identify railcars and detect defects in freight wagons., Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Multi-interest dual-adaptive unscented Kalman Filter with online parameter optimization for nonlinear battery modeling

  • Rasheed, Muhammad Babar|||0000-0002-9911-0693
  • Llamazares Llamazares, Ángel|||0000-0001-8273-5163
  • Gutiérrez Moreno, Rodrigo
  • Ocaña Miguel, Manuel|||0000-0002-8875-1866
  • Revenga de Toro, Pedro Alfonso|||0000-0003-2550-5972
Reliable driving range prediction is a key requirement for modern electric mobility. However, it remains challenging due to the complex electrochemical dynamics of lithium-ion batteries and their sensitivity to varying operating conditions. To address these challenges, this study proposes a novel framework integrating an enhanced second-order RC equivalent circuit model with a multi-criterion adaptive Unscented Kalman Filter algorithm (i-DAUKF) for improved state estimation accuracy. Firstly, the enhanced RC model is developed to optimize the battery parameter selection using a moving window average algorithm, while incorporating online parameter optimization to dynamically update measurement and predictive coefficients. The i-DAUKF algorithm is designed to better capture nonlinear battery behavior under operational thermal fluctuations and degradation processes, ensuring robust estimation fidelity across diverse operating conditions. Unlike conventional filtering approaches that rely on static tuning, the proposed method continuously refines model parameters, significantly improving adaptability and robustness. The novel contributions include: (i) integration of a refined second-order RC model with adaptive parameter calibration, (ii) dynamic optimization of predictive and measurement gains for sustained stability, and (iii) demonstration of application-specific improvements across transient and steady-state phases. Simulation results validate the effectiveness of the framework, achieving SOC estimation errors of 0.043 and 0.714% (RMSE), and voltage errors of 0.320 and 1.42% (RMSE), respectively. Comparative evaluations further reveal that i-DAUKF consistently achieves the lowest estimation errors (SOC RMSE <2,5%, voltage 1%?3%) with negligible computational cost, thereby establishing its superiority in accuracy, robustness, and real-time efficiency for nonlinear battery modeling under dynamic driving conditions., European Union, Ministerio de Ciencia, Innovación y Universidades, Comunidad de Madrid
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Leveraging Driver Attention for an End-to-End Explainable Decision-Making from Frontal Images

  • Araluce Ruiz, Javier
  • Bergasa Pascual, Luis Miguel|||0000-0002-0087-3077
  • Ocaña Miguel, Manuel|||0000-0002-8875-1866
  • Llamazares Llamazares, Ángel|||0000-0001-8273-5163
  • López Guillén, María Elena
Explaining the decision made by end-to-end autonomous driving is a difficult task. These approaches take raw sensor data and compute the decision as a black box with large deep learning models. Understanding the output of deep learning is a complex challenge due to the complicated nature of explainability; as data passes through the network, it becomes untraceable, making it difficult to understand. Explainability increases confidence in the decision by making the black box that drives the vehicle transparent to the user inside. Achieving a Level 5 autonomous vehicle necessitates the resolution of that challenging task. In this work, we propose a model that leverages the driver’s attention to obtain explainable decisions based on an attention map and the scene context. Our novel architecture addresses the task of obtaining a decision and its explanation from a single RGB sequence of the driving scene ahead. We base this architecture on the Transformer architecture with some efficiency tricks in order to use it at a reasonable frame rate. Moreover, we integrate in this proposal our previous ARAGAN model, which obtains SOTA attention maps, to improve the performance of the model thanks to understand the sequence as a human does. We train and validate our proposal on the BDD-OIA dataset, achieving on-pair results or even better than other state-of-the-art methods. Additionally, we present a simulation-based proof of concept demonstrating the model’s performance as a copilot in a close-loop vehicle to driver interaction., Agencia Estatal de Investigación
e_Buah Biblioteca Digital Universidad de Alcalá
Publicación Artículo científico (JournalArticle)

Context-aware state estimation in battery management systems: leveraging nonlinear dynamics with physics-guided parameter identification

  • Rasheed, Muhammad Babar|||0000-0002-9911-0693
  • Llamazares Llamazares, Ángel|||0000-0001-8273-5163
  • Gutiérrez Moreno, Rodrigo
  • Ocaña Miguel, Manuel|||0000-0002-8875-1866
  • Revenga de Toro, Pedro Alfonso|||0000-0003-2550-5972
Black accurate remaining range estimation remains a critical issue to promote plug-in and electric vehicle adoption, primarily due to underlying uncertainties in voltage and current-dependent state estimation. To overcome these challenges, the proposed work introduces a novel framework for range estimation while integrating an enhanced equivalent circuit model with a physics-guided temperature-compensated Extended Kalman Filter algorithm. Firstly, comprehensive mathematical models are developed and validated that integrate: (i) proposed enhanced 3rd-order equivalent circuit modeling (p-eTECM) with control parameter optimization, (ii) data-and-model-driven parameter identification using Coulomb counting and voltage scaling analysis, (iii) comprehensive sensitivity analysis to rank important parameters to improve accuracy, and (iv) application-specific model selection criteria based on performance trade-offs. However, unlike existing frameworks that incorporate higher-order RC models that are universally superior, the proposed work identifies that model selection should be application-dependent for different battery management functions. The novel contributions include: parameter & voltage optimization from the pack-level, while systematically eliminating voltage bias through online parameter optimization, and developing a comprehensive sensitivity analysis algorithm to validate the improvements. The proposed framework demonstrates that parameter calibration is more crucial with capacity correction and voltage scaling, to eliminate systematic biases that render models impractical. This study further reveals that 3rd-order model outperforms in voltage prediction (8.3 % improvement) while the 2nd-order model provides better SOC tracking (13 % improved accuracy), establishing clear application-specific selection criteria. Key results demonstrate that both models have achieved excellent performance in terms of SOC errors (<0.2 %), and range accuracy (155?170 km) with real-time computational efficiency, validating the practical applicability for diverse battery management applications while providing a systematic methodology for future battery modeling research., Agencia Estatal de Investigación