IET Intelligent Transport Systems最新文献

The structural integrity of railway fasteners is a critical determinant of track safety, load transfer stability, and long-term infrastructure reliability. In practice, fastener loosening often develops gradually and manifests through subtle geometric, depth, and mechanical changes that are difficult to capture using single-modality or label-dependent inspection methods. Existing deep learning approaches largely rely on manually annotated datasets, isolated visual cues, or single-sensor inputs, which limit their robustness under real-world variability, noise, and data scarcity. To address these challenges, this paper proposes TrinityNet, a cross-modality generative reasoning framework for zero-label fastener tightness evaluation. TrinityNet jointly exploits RGB imagery, monocular depth estimation, and 3D mesh representations to capture complementary structural and geometric information without requiring manual annotations. A self-supervised dual-evaluator architecture independently validates each modality through adversarial and contrastive learning, improving training stability and cross-modal consistency. Experimental results demonstrate a 99.64% reduction in discriminator loss, an 80.55% decrease in contrastive error, and a 409.67% improvement in generative consistency. Fastener tightness is quantitatively assessed using a multi-metric diagnostic scheme incorporating fatigue life, principal component analysis–based degradation indices, cyclic load response, and energy strain measures. The framework reliably distinguishes tight fasteners from loose, enabling interpretable and actionable maintenance decisions. Owing to its label-free learning, multimodal robustness, and real-time applicability, TrinityNet provides a practical and scalable solution for autonomous railway fastener integrity monitoring.

This paper introduces MOMSATC, an innovative multi-objective, multi-step adaptive traffic signal control framework grounded in the principles of model predictive control. MOMSATC is specifically designed to address complex, high-dimensional optimisation challenges, including the mitigation of pedestrian and vehicle safety risks alongside delay management. The framework first establishes a hybrid safety evaluation model to comprehensively assess conflicts involving vulnerable road users, providing input to a multi-task learning model that predicts safety and delay outcomes. Safety risks are translated into a quantifiable monetary cost equivalent using a willingness-to-pay approach that considers long-term health and socio-economic impacts. The overarching aim of MOMSATC is to support an interpretable decision process that can represent objective priorities in a transparent manner. By integrating predictive modelling with a structured optimisation procedure, the framework allows pedestrian safety to be prioritised while maintaining a balance between vehicle safety and overall operational efficiency. A case study demonstrates the efficacy of MOMSATC, achieving significant reductions in safety risks for both pedestrians and vehicles, with moderate trade-offs in delay, underscoring its potential to achieve a safety-orientated urban transport system.

Vertical alignment design for heavy-haul railways not only profoundly affects construction investment but also directly influences train energy consumption. Under complex and interdependent constraints, designers must reconcile the conflict between these two objectives; however, most existing methods require the number of vertical points of intersection (VPIs) to be predetermined and suffer severe efficiency losses as constraints become more intricate. To address these challenges, this study, for the first time, employs an NSGA-II framework enhanced with struct-based encoding to achieve bi-objective optimization of construction cost and energy consumption. The model requires no preset VPIs and naturally satisfies most constraints. Under the given constraints, it can produce schemes that outperform manually designed routes in both construction cost and train energy consumption, achieving reductions of 4.09%–6.41% and 2.37%–5.92%, respectively, under different heavy-to-light ratio conditions. while also offering guidance for selecting technically optimal ratios in future heavy-haul railway design.

With the rapid development of intelligent cockpits, personalized recommendations have become crucial for achieving high-level cognitive intelligence in cockpit systems. The core goal is to mine the behavioural history of drivers and passengers to provide tailored, proactive interactions based on environmental conditions and user preferences. Current systems mainly rely on rules or limited offline data, focusing on specific functions or scenarios, which lack global capabilities and struggle with multi-task collaboration, leading to inaccuracies and limited flexibility in personalized recommendations. Large language models (LLMs), with their powerful general-purpose understanding capabilities, have demonstrated significant advantages in reasoning about complex user intentions and enhancing interaction recommendation performance. However, LLMs have not been applied to cockpit personalized interaction recommendations. To bridge this gap and effectively balance the complexity of cockpit systems under highly diverse multi-task and personalized requirements, this paper proposes an innovative two-stage recommendation framework, PC-LLMRec, specifically designed for customized recommendations in intelligent cockpits. The framework employs full-parameter baseline model optimization and personalized adapter construction to achieve general recommendations in the cloud and personalized adjustments on the vehicle end. This allows precise capture and interpretation of both common behavioural patterns and individualized user needs, enabling cross-scenario, multi-task proactive recommendation. To enhance the adaptability of PC-LLMRec, this paper also constructs an instruction-following dataset tailored to proactive cockpit interaction recommendations. This dataset includes extensive user interaction context and real recommendation labels, ensuring effective fine-tuning between global recommendations and personalized services. Extensive experimental results demonstrate that PC-LLMRec excels in accuracy and adaptability across various recommendation scenarios, outperforming existing context-learning-based methods, retrieval-augmented prompt strategies, and other state-of-the-art models.

Pedestrian trajectory prediction based on computer vision technology is crucial for automatic driving systems and robot vision. This study proposes the use of deep CoordConv with autoencoders for the high-precision prediction of pedestrian trajectories and endpoints in real-time. First, an autoencoder-based model combines with CoordConv using a past trajectory encoder, endpoint decoder and future trajectory decoder to enhance the coordinate features. Second, the proposed model predicts the possible endpoints and generates the trajectory from the start predicted position to each endpoint to overcome the multi-modality problem. Finally, in extensive experiments, the proposed model for short-term, long-term and endpoint predictions outperformed conventional RNN-based models.

Vehicular ad hoc networks (VANETs) are fundamental to intelligent transportation systems (ITS), enabling secure and low-latency vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communication. Conditional privacy-preserving authentication (CPPA) is essential for safeguarding message integrity and anonymity, yet traditional ECC- and pairing-based CPPA schemes are both computationally intensive and vulnerable to quantum attacks. Although lattice-based CPPA (L-CPPA) schemes offer post-quantum resistance and batch verification, their reliance on a single roadside unit (RSU) introduces verification bottlenecks and a single point of failure in dense traffic scenarios. To overcome these limitations, we propose a multi-aggregator lattice-based CPPA (MA-LCPPA) framework that distributes verification tasks across cooperating RSUs and integrates a (k,n)-threshold traceability mechanism. This design significantly reduces verification delay, improves scalability and enhances fault tolerance while maintaining conditional privacy and post-quantum security. Formal analysis demonstrates unforgeability, traceability and resilience against replay, impersonation, and collusion attacks under the hardness of the ISIS problem. Simulation results confirm that MA-LCPPA reduces verification delay by over 50% and lowers RSU computation costs, with minimal communication overhead, making it a scalable and quantum-secure solution for next-generation vehicular networks.

To improve the management and operational efficiency of Intelligent Transportation Systems (ITS), address the nonlinear complexity of short-term traffic flow, mitigate the issue of significant noise in traffic flow datasets, and tackle the challenges in determining parameters for Support Vector Machines (SVM) and Long Short-Term Memory (LSTM) networks, this paper proposes a short-term traffic flow prediction model based on Variational Mode Decomposition (VMD) and Least Squares Support Vector Machine (LSSVM) integrated with an attention mechanism. Multiple intrinsic mode functions (IMFs) decomposed by VMD are input into the LSSVM model, and the parameters and weights of the model are automatically adjusted using the attention mechanism. Experimental results on the Italian highway traffic flow dataset show that the prediction accuracy of the VMD-LSSVM-Attention model is improved by an average of about 38.6% compared with the traditional VMD-SVM, VMD-LSTM-Attention, VMD-LSSVM and LSSVM-Attention models, and the model is more stable. Furthermore, in generalisation validation experiments on the Rotterdam and Madrid datasets, the model improved prediction accuracy by 5.17% to 20.97% compared to the best-performing advanced models. This model provides a prediction method for the traffic flow prediction module in the intelligent transportation system (ITS) architecture.

This paper presents an integrated hybrid reinforcement learning–model predictive control (RLMPC) framework for autonomous highway systems, unifying macroscopic traffic flow regulation and microscopic on-ramp merging control. At the macroscopic level, a ramp metering (RM) controller based on a data-driven model predictive control (MPC) formulation using second-order Q-learning is implemented in the METANET environment on a benchmark three-segment freeway without the need for explicit traffic models. The RLMPC RM learns optimal flow regulation directly from closed-loop data, achieving enhanced system performance, constraint satisfaction and smooth control compared to common RM algorithms such as ALINEA, MPC and deep RL. At the microscopic level, an RLMPC merging controller manages autonomous on-ramp manoeuvres in which an ego vehicle enters the mainline approximately 160 m before the merge point and completes the manoeuvre 50 m downstream while interacting with surrounding vehicles. In this phase, when a collision risk arises, the MPC takes control; otherwise, the reinforcement learning (RL) policy operates, combining model-based safety with learning-based efficiency and yielding superior overall performance. Evaluations under varied traffic conditions show that implementing RM at the macroscopic level significantly improves microscopic on-ramp merging performance. Relative to the no-RM baseline, the framework achieves a 34.5% reduction in merge time under slow traffic conditions, eliminates collision events and moderately enhances overall efficiency and driving comfort.

To address the issues of missed detections, false positives and feature degradation in road debris detection caused by small and irregular targets, this study proposes a novel framework integrating multi-scale feature fusion and dynamic feature enhancement mechanisms. It also constructs a dedicated road debris dataset to fill the gap in public benchmark datasets in this field. Firstly, a cross-layer connection-optimized feature fusion network is designed in the neck network, addressing the limitation of insufficient fusion of shallow and deep features in existing methods, realizing efficient linkage between shallow texture features and deep semantic information, and significantly improving the detection capability for small targets. Secondly, a context-aware anchor attention module integrating reparameterized convolution and adaptive weight allocation is embedded into the backbone network. Compared with traditional fixed receptive field convolution, it can dynamically enhance target features and suppress background interference, effectively solving the problem of feature degradation in complex environments. Thirdly, an improved spatial pyramid fast pooling module based on global pooling and Ghost convolution is proposed, overcoming the defect of prone detail loss in traditional max-pooling and preserving key information of small-sized road debris to the greatest extent. Finally, a weighted fusion loss function integrating corner distance loss, focal loss, cross-scale correlation loss and CIoU loss is designed, breaking the limitation of insufficient attention to irregular targets in a single loss function and enhancing the model's adaptability to complex scenes. Experimental results show that the framework outperforms existing mainstream methods in road debris detection scenarios, achieving a precision of 91.5%, a recall of 82.0% and an mAP50 of 88.7%.

Cooperative adaptive cruise control, also known as vehicular platooning, has gained significant interest for its ability to enhance fuel efficiency and comfort in vehicle operations. This study proposes novel control strategies for vehicular platooning based on long short-term memory (LSTM) neural networks. By learning temporal dependencies in vehicle behaviour, the proposed LSTM-based controllers improve string stability within the platoon, particularly under varying velocity patterns of the lead vehicle. Two distinct frameworks are investigated: centralized and decentralized control models. The centralized model makes use of the states of all vehicles within the platoon, whereas the decentralized model focuses on the states of only a limited number of preceding vehicles. Simulation experiments demonstrate that both the centralized and decentralized LSTM controllers significantly outperform traditional, non-LSTM-based controllers in minimizing cumulative inter-vehicle error. This study contributes a novel controller training methodology that integrates LSTM-based architectures with optimal control principles, offering improved adaptability and flexibility in real-time platoon management.