Convolutional Neural Networks for Accurate Measurement of Train Speed

2025-08-23

AI Summary

To address the degradation of train speed estimation accuracy under complex operational conditions—such as wheel slip—in modern railway systems, this paper proposes a multi-branch convolutional neural network (CNN) that synergistically integrates 1D-CNN (for temporal feature extraction) and 2D-CNN (for spatiotemporal pattern modeling). The method achieves highly robust speed estimation without relying on explicit vehicle dynamics models. Compared to conventional single-branch CNNs and adaptive Kalman filtering, the proposed model demonstrates significantly improved accuracy and stability on simulated data featuring activated slip protection. Its key innovation lies in the joint modeling of multimodal temporal and spatial features, effectively mitigating the strong model dependency and poor adaptability to varying operating conditions inherent in traditional approaches. Experimental results confirm that the method substantially enhances speed measurement reliability, providing critical technical support for intelligent train control and operational safety.

Technology Category

• Intelligent Robots: State Estimation
• Machine Learning: Multimodal Learning
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

In this study, we explore the use of Convolutional Neural Networks for improving train speed estimation accuracy, addressing the complex challenges of modern railway systems. We investigate three CNN architectures - single-branch 2D, single-branch 1D, and multiple-branch models - and compare them with the Adaptive Kalman Filter. We analyse their performance using simulated train operation datasets with and without Wheel Slide Protection activation. Our results reveal that CNN-based approaches, especially the multiple-branch model, demonstrate superior accuracy and robustness compared to traditional methods, particularly under challenging operational conditions. These findings highlight the potential of deep learning techniques to enhance railway safety and operational efficiency by more effectively capturing intricate patterns in complex transportation datasets.

Problem

• Improving train speed estimation accuracy using CNNs
• Comparing CNN architectures with Adaptive Kalman Filter
• Enhancing railway safety through deep learning techniques

Innovation

• CNN architectures for train speed estimation
• Multiple-branch model outperforms traditional methods
• Deep learning enhances railway safety efficiency