This study develops an integrated dynamic routing and off-block (IDRO) optimization framework for real-time airport surface operations. Instead of relying on complex taxi time prediction models, the proposed approach employs a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">dynamic route impedance</i> formulation as a practical heuristic to capture real-time congestion conditions on the airport surface. The impedance integrates domain-specific operational factors, including free flow taxi time, turning penalties, and potential aircraft conflicts, enabling real-time route evaluation and decision-making without extensive simulation. Building upon this impedance representation, the IDRO framework coordinates both taxi routing and off-block release timing under a real-time First-Come-First-Served (FCFS) principle. The approach was implemented and tested using a high-fidelity cellular automata simulator calibrated with operational data from Beijing Capital International Airport. Compared with real-world baseline operations, the proposed framework reduced average taxi time, flight delay, and taxi conflicts by 10.43%, 24.27%, and 31.82%, respectively. A comparative study with another similar approach in the literature also reveals the advantage of the proposed strategy. These results demonstrate the framework’s computational efficiency, consistent performance across scenarios, and strong potential for real-world deployment in large-scale airport environments.

Traffic trajectory prediction plays a crucial role in ensuring the safety and efficiency of autonomous driving systems. With increasing traffic complexity, accurate and real-time prediction of surrounding vehicle trajectories has become a major challenge. Existing methods often define trajectory prediction as a static task, predicting future trajectories at once based solely on historical states. However, this approach leads to significant long-term prediction errors and struggles to capture potential hazards in complex, dynamic traffic conditions. In this paper, we propose a lightweight end-to-end trajectory prediction model integrating a long-term guidance module and a short-term feedback mechanism. The long-term guidance module predicts the final driving goal that remains unchanged over time, guiding the vehicle’s eventual position. The short-term feedback mechanism captures the dynamic driving environment, continuously updating the model with immediate changes in vehicle interactions and road conditions to avoid unexpected dangers. At each time step, our approach leverages accumulated historical information and incorporates both short-term and long-term future information to iteratively predict multi-modal trajectories, effectively reducing long-term prediction errors. Additionally, our method introduces the Mamba module, which enhances real-time prediction performance and reduces historical information loss during long-term prediction. Experiments on the Argoverse 1 and Argoverse 2 datasets show that our method, with fewer parameters and higher inference speed, achieves results comparable to or surpassing previous state-of-the-art methods.

The effectiveness of existing testing methods is under scrutiny due to significant limitations, notably the discrepancy between action distributions and actual distributions caused by inadequate environmental understanding. Additionally, the lack of a scheme prioritizing fidelity while coordinating testing efficiency results in considerable divergence between test and natural scenarios. To address these issues, this paper proposes a Directed Graph Reinforcement Learning approach with action constraint optimization (DGRL) to generate critical scenarios, balancing efficiency and fidelity. By incorporating directed graph convolutional networks, the model encodes environmental states within the observation interval, providing spatiotemporal insights and reducing action distribution discrepancies. Furthermore, it constructs an unbiased estimation reward considering action constraints by sampling action confidence intervals, filtering out distorted actions, thus balancing efficiency and fidelity. DGRL was trained using the highD dataset, demonstrating robust acceleration performance, with interquartile range (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$IQR$</tex-math> </inline-formula>) of 3.1 and first quartile (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q_{1}$</tex-math> </inline-formula>) of 75.6 in the acceleration ratio distribution, representing the bandwidth and baseline, respectively. The model also achieved high fidelity, with scenario discrepancies compared to natural scenarios reduced by 85.4% (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q_{1}$</tex-math> </inline-formula>) and 46.5% (<inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$IQR$</tex-math> </inline-formula>) relative to GAIL, which considers the rationality of driving behavior. Here, <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$Q_{1}$</tex-math> </inline-formula> represents the baseline of scenario discrepancy, and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$IQR$</tex-math> </inline-formula> denotes the distribution bandwidth of scenario discrepancy. Additionally, there was 69.7% reduction in the upper limit of the 95% confidence interval, indicating a significant decrease in maximum scenario discrepancy. Deployment on an intelligent connected hardware-in-the-loop testing platform validated DGRL’s effectiveness and applicability in real-world.

The State Space Model (SSM) has been a growth of interest in computer vision due to its long-term dependency modeling with linear complexity. Despite massive endeavor, it has not been extensively explored in intelligent transportation system (ITS) yet. In this paper, we propose State Space Model-based object Detector (<italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">SSM-Det</i>), that is meticulously curated with Direction-aware Visual State Space Encoder (D-VSSE). Specifically, it customizes multi-path pixel exchange and patch re-arrangement via four-direction scanning mechanism, promoting for information communication. To bridge the information bottleneck across high-low level, we further design Split-Fusion (SF) and Skip-Connection (SC) modules for contextual feature propagation before decoding: SF performs multi-channel semantic separation and re-weighting in global-local scope, while SC is responsible for cross-layer feature interaction in a cascaded manner. Empirical studies is conducted on both VisDrone2019-DET and SEU_PML benchmarks, and our proposed SSM-Det reports the state-of-the-art performance against all counterparts by a substantial margin, while maintaining the real-time inference speed. We hope this work contributes to the in-depth investigation of SSM-based detector for intelligent transportation applications. The code is available at <uri xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">https://buaawjq.github.io/SSM-Det/</uri>.

In complex and dynamic maritime environments, the uncertain navigation intent of target ships (TS) poses a significant challenge for the collision avoidance strategies of unmanned surface vehicles (USVs). To address above issue, an autonomous collision avoidance method that integrates recursive Bayesian estimation, a masked attention mechanism, and the soft actor-critic (SAC) algorithm is proposed in this paper. Specifically, the recursive Bayesian model is employed to estimate the navigation intentions of TS in real time, for enabling the construction of intent-aware state representations that enhance the modeling capability for uncertain target behaviors. In addition, a maximum target count is predefined, and the masked attention mechanism is incorporated into the actor and critic networks of SAC to resolve the mismatch between the fixed input dimensionality of neural networks and the dynamically varying number of TS. Simulation demonstrates that the proposed method outperforms traditional collision avoidance methods and other state-of-the-art deep reinforcement learning (DRL)-based strategies in terms of key performance metrics such as collision avoidance success rate, minimum distance, and average risk. Furthermore, real-field experiments validate the proposed method’s adaptability to real-world maritime environments and its potential for practical deployment.

Point cloud semantic segmentation, which involves assigning a category for each point, is a crucial task in autonomous driving and intelligent transportation systems. Due to the inherently unordered and irregular nature of point clouds, learning robust features that accurately capture real-world distributions from point coordinates and other attributes remains challenging. Following the pioneering work of PointNet, current 3D deep neural networks process point coordinates alongside other attributes without fully exploiting the implicit class prior information embedded in spatial information. In this work, we first conduct a pilot study to evaluate how current 3D networks utilize point coordinates and validate the presence of implicit class priors within them. Subsequently, we design a robust Position-to-Physics (P2P) fusion strategy that learns adaptive weights to dynamically incorporate implicit class priors present in point coordinates into point features. Moreover, we design a dual-branch network architecture and propose a triplet loss to further enhance the adaptive fusion process. Extensive experiments demonstrate that decoupling position attributes from physics attributes facilitates the extraction and utilization of implicit class priors. Our proposed modules consistently improve segmentation performance across various networks and datasets, demonstrating their generalizability and effectiveness.

Transportation Networks (TNs) play a critical role in economic and social systems, yet the dynamic nature and inherent heterogeneity of TN data pose challenges for Dynamic Knowledge Integration (DKI). Traditional approaches for matching entities from different knowledge bases often struggle with the complexity and diversity of TN data, which varies across systems and sources such as traffic sensors and GPS devices. To address this issue, this paper proposes a novel Multi-Objective Genetic Programming assisted Stochastic Deep Reinforcement Learning (MOGP-SDRL) for DKI in TNs. Unlike existing methods, the proposed framework combines SDRL and MOGP to achieve superior efficiency, accuracy and adaptability in handling heterogeneous TN data. First, a novel SDRL framework is designed to automate and optimize the selection of Similarity Features (SFs) for entity matching. This framework incorporates a probabilistic action selection mechanism, which enhances exploration during the SF selection process. Second, a novel MOGP is presented to construct high-quality, diverse SFs by exploring non-dominated feature ensembles, enhancing both accuracy and adaptability in matching results, leading to more accurate and adaptable matching results compared to conventional methods. Lastly, new approximate evaluation metrics are developed to assess alignment quality without relying on predefined entity alignments, guiding the optimization process. Experimental evaluations on OAEI’s knowledge graph (KG) dataset and five pairs of real-world TN dataset demonstrate the effectiveness of the MOGP-SDRL framework, which consistently produces high-quality matching results and achieves significant improvements in both accuracy and robustness over existing approaches.

Smart transportation networks face increasing demands for efficiency and sustainability. This study presents a reinforcement learning approach that optimizes eco-driving strategies for connected and automated vehicles (CAVs) in urban environments, with a particular application to healthcare logistics. Specifically, we propose a novel approach using reinforcement learning, specifically a twin delayed deep deterministic policy gradient (TD3) algorithm, to dynamically optimize CAV trajectories at signalized intersections. The proposed healthcare eco-driving trajectory optimization (TD3-HETO) model incorporates real-time traffic conditions, signal timing information, and healthcare urgency levels to generate optimal acceleration profiles. The reward function is designed to balance energy efficiency, traffic flow, safety, comfort, and healthcare delivery timeliness. Additionally, the model introduces a dynamic exploration strategy that adapts to healthcare task urgency, enabling efficient balancing between energy consumption and delivery timelines. Experimental results show that TD3-HETO reduces energy consumption by up to 28.7% compared to baseline methods while improving average speeds by 3.7% for urgent healthcare deliveries. The model achieves superior safety performance with 98.7% of time steps showing zero conflicts, compared to 95.3% for the best baseline. TD3-HETO also demonstrates remarkable adaptability to varying traffic demands and signal timings, maintaining consistent performance even at high traffic volumes. This research contributes to developing intelligent transportation systems to enhance environmental sustainability and healthcare accessibility in smart cities, potentially improving patient outcomes and operational efficiency in urban healthcare logistics.