27 April 2026 in IEEE Transactions on Instrumentation and Measurement
The Catenary Support Component (CSC) is a critical element in ensuring the safe operation of electrified railway systems. In recent years, the use of UAVs for railway maintenance and inspection has steadily increased, making them an increasingly important technological tool. However, UAVs face several challenges during railway inspections. These challenges include energy consumption constraints, which limit their prolonged use, as well as complex backgrounds, structural occlusions, and varying component scales that hinder detection accuracy. To address these challenges, a low-energy consumption, graph-guided spike-based brain-inspired neural network algorithm (GSINet) was proposed. First, a novel backbone called the spatiotemporal multiscale spike fusion network was designed. This network leverages spatiotemporal spike encoding for feature modeling and enables cross-scale feature fusion. Temporal and channel attention mechanisms, along with SpikeASPP technology, were used to enhance the network’s ability to handle diverse scales and complex environmental conditions. Next, a Spike-inspired detection head was designed with a position guidance module (PGM) based on graph convolutional networks. The PGM utilizes prior category relationships, integrating both semantic and geometric information to mitigate challenges such as occlusion and multi-component measurement. Finally, a spiking network architecture was implemented using the improved Leaky Integrate-and-Fire (IB-LIF) Neuron. This reduces energy consumption by exploiting sparse computations and an event-driven mechanism, while ensuring high detection accuracy. Extensive experiments on a CSC dataset captured by UAVs demonstrate that the model achieves superior measurement performance while consuming significantly lower energy.
14 June 2026 in SSRN Electronic Journal
As a critical subsystem in the operation and maintenance of electrified railways, the intelligent structural health monitoring (SHM) of catenary support components faces several key engineering challenges: (1) randomness and diversity of structural damage modes make it impractical to construct comprehensive supervised datasets; (2) the complex geometric topology of catenary support assemblies leads to information loss of subtle features during deep network extraction; (3) current reconstruction models, lacking normative semantic constraints, tend to replicate input defect features, thus leading to ineffective structural repair of abnormal catenary regions. To address these issues, we propose a Global State-aware Clustering-guided Masked Autoencoder (GSC-MAE) for robust unsupervised anomaly detection of catenary support structures. First, a Global State Aggregation (GSA) backbone leverages State Space Modeling (SSM) to enhance global topological perception, mitigating micro-structural feature loss in complex catenary components. Second, clustering-driven adaptive masking targets anomalous catenary regions, enforcing restoration from normal structural contexts and blocking defect propagation. Additionally, a SCGFormer decoder utilizes strided memory flow to prevent information forgetting, enhancing reconstruction fidelity of intricate catenary topologies. Extensive UAV-based experiments yield 99.4% I-AUROC and 98.5% I-AP, surpassing state-of-the-art methods. This framework provides an efficient solution for automated catenary structural integrity assessment, promoting intelligent infrastructure maintenance.
1 January 2026 in IEEE Transactions on Transportation Electrification