Sistema Escalable de Detección de Vehículos para Infraestructuras Inteligentes

• Borau Bernad, Javier ^[1] ; Armingol Moreno, José María ^[1] ; Sanchis de Miguel, Araceli ^[1]
  1. [1] Universidad Carlos III de Madrid

     Universidad Carlos III de Madrid

     Madrid, España

     
• Localización: Jornadas de Automática, ISSN-e 3045-4093, Nº. 46, 2025
• Idioma: español
• DOI: 10.17979/ja-cea.2025.46.12076
• Títulos paralelos:
  • Scalable Vehicle Detection System for Intelligent Infrastructures
• Enlaces
  • Texto completo
• Resumen
  • español

    El aumento de la población en las ciudades ha creado la necesidad de gestionar de una forma eficiente los sistemas de transporte para mejorar la seguridad vial, optimizar el tráfico y reducir su impacto medioambiental. Las infraestructuras inteligentes equipadas con tecnologías de detección emergen como la principal solución para la monitorización del tráfico, pero enfrentan desafíos relacionados con costes, precisión y complejidad de instalación. En este contexto, este artículo presenta un sistema escalable de detección tridimensional de vehículos que integra tres modalidades de detección: monocular, LiDAR y multimodal (LiDAR y cámara RGB). El sistema propuesto selecciona automáticamente el modo de operación en función de los sensores instalados en cada infraestructura. Los resultados obtenidos muestran que esta solución modular permite optimizar el equilibrio entre coste y precisión para facilitar la implementación progresiva según las necesidades específicas de cada entorno urbano.

  • English

    The growing urban population has increased the need for efficient transportation systems that enhance road safety, optimize traffic flow and reduce environmental impact. Intelligent infrastructures equipped with sensing technologies have emerged as a key solution for traffic monitoring; however, they still face challenges related to cost, accuracy and installation complexity. This article presents a scalable 3D vehicle detection system that supports three detection modes: monocular, LiDAR and multimodal (LiDAR combined with an RGB camera). The proposed system automatically selects the most suitable mode based on the sensors available in each infrastructure. Experimental results show that this modular approach effectively balances cost and performance, enabling flexible and progressive deployment according to the specific needs of each urban environment.

• Referencias bibliográficas
  • Arnold, E., Al-Jarrah, O. Y., Dianati, M., Fallah, S., Oxtoby, D., Mouzakitis, A., 2019. A survey on 3d object detection methods for autonomous...
  • Borau Bernad, J., 2024. RoadVision3D. https://github.com/jborau/RoadVision3D.
  • Borau Bernad, J., Ramajo-Ballester, Á., Armingol Moreno, J. M., 2024. Three-dimensional vehicle detection and pose estimation in monocular...
  • Dong, Q., Zhou, Z., Qiu, X., Zhang, L., 2025. A survey on self-supervised monocular depth estimation based on deep neural networks. IEEE Transactions...
  • Geiger, A., Lenz, P., Urtasun, R., 2012. Are we ready for autonomous driving? The KITTI vision benchmark suite. 2012 IEEE Conference on Computer...
  • Huang, K., Shi, B., Li, X., Li, X., Huang, S., Li, Y., 2024. Multi-modal sensor fusion for auto driving perception: A survey. URL: https://arxiv.org/abs/2202.02703
  • Li, Z., Jia, J., Shi, Y., 2024. MonoLSS: Learnable sample selection for monocular 3d detection. In: 2024 International Conference on 3D Vision...
  • Liu, X., Xue, N., Wu, T., 2021. Learning auxiliary monocular contexts helps monocular 3d object detection. URL: https://arxiv.org/abs/2112.04628
  • Liu, Z., Wu, Z., Tóth, R., 2020. SMOKE: Single-stage monocular 3d object detection via keypoint estimation. URL: https://arxiv.org/abs/2002.10111
  • MMDetection3D Contributors, 2020. MMDetection3D: OpenMMLab next-generation platform for general 3D object detection. https://github.com/open-mmlab/mmdetection3d
  • Owais, M., 2024. Deep learning for integrated origin–destination estimation and traffic sensor location problems. IEEE Transactions on Intelligent...
  • Sindagi, V. A., Zhou, Y., Tuzel, O., 2019. MVX-Net: Multimodal voxelnet for 3d object detection. In: 2019 International Conference on Robotics...
  • Wang, Y., Mao, Q., Zhu, H., Deng, J., Zhang, Y., Ji, J., Li, H., Zhang, Y., 2023. Multi-modal 3d object detection in autonomous driving: A...
  • Yan, Y., Mao, Y., Li, B., 2018. SECOND: Sparsely embedded convolutional detection. Sensors 18 (10). URL: https://www.mdpi.com/1424-8220/18/10/3337...
  • Yu, F., Wang, D., Shelhamer, E., Darrell, T., 2018. Deep layer aggregation. In: Proceedings of the IEEE Conference on Computer Vision and...
  • Yu, H., Luo, Y., Shu, M., Huo, Y., Yang, Z., Shi, Y., Guo, Z., Li, H., Hu, X., Yuan, J., Nie, Z., June 2022. DAIR-V2X: A large-scale dataset...
  • Zanella, A., Bui, N., Castellani, A., Vangelista, L., Zorzi, M., 2014. Internet of Things for smart cities. IEEE Internet of Things Journal...
  • Zhang, Y., Carballo, A., Yang, H., Takeda, K., 2023. Perception and sensing for autonomous vehicles under adverse weather conditions: A survey....
  • Zhou, Y., Tuzel, O., 2017. VoxelNet: End-to-end learning for point cloud based 3d object detection. URL: https://arxiv.org/abs/1711.06396
  • Zimmer, W., Ercelik, E., Zhou, X., Ortiz, X. J. D., Knoll, A., 2022. A survey of robust 3d object detection methods in point clouds. URL:...