Trajectory data has become a key resource for automated map inference due to its low cost, broad coverage, and continuous availability. However, uneven trajectory density often leads to fragmented roads in sparse areas and redundant segments in dense regions, posing significant challenges for existing methods. To address these issues, we propose DGMap, a dual-decoding framework with global context awareness, featuring Multi-scale Grid Encoding, Mask-enhanced Keypoint Extraction, and Global Context-aware Relation Prediction. By integrating global semantic context with local geometric features, DGMap improves keypoint detection accuracy to reduce road fragmentation in sparse-trajectory areas. Additionally, the Global Context-aware Relation Prediction module suppresses false connections in dense-trajectory regions by modeling long-range trajectory patterns.Experimental results on three real-world datasets show that DGMap outperforms state-of-the-art methods by 5% in APLS, with notable performance gains on trajectory data from the Didi Chuxing platform.

Due to the high coverage and low cost nature of trajectory data, an increasing number of works have utilized trajectory data to infer maps. Nevertheless, limited by the sparse trajectories in some areas and intermingled trajectories on parallel roads, the existing inferring methods still face a high missed detection rate of the roads. In view of that, we propose a Complementarity and Disparity-aware Map Inference Framework, called CDMap, consisting of grid dual feature extraction, contextual road difference-embedded grid representation, dual feature complementary network-based road topology prediction and parallel roads disparity-enhanced model optimization. To improve the prediction accuracy of the roads in areas with sparse trajectories, we extract point-wise features and segment-wise features separately for the grids, then design a dual feature complementary network to adaptively model the importance of both types of features in different road scenarios. Further, to proliferate the detection accuracy of parallel roads, we incorporate the contextual roads' differences between parallel roads into grid representations, then put forward a parallel roads disparity-enhanced model optimization strategy. Extensive comparative experiments conducted on three real-world datasets demonstrate the superiority of CDMap over the state-of-the-art methods, especially by achieving the most significant reduction in missed detection rate (30.23%) on the trajectory data collected from DidiChuxing platform.

Abstract Truck-cargo matching is one of the core tasks of online freight platforms, where the primary objective is to optimally assign each cargo task to the most suitable truck. The existing matching strategies seek to maximize total transported weight of cargoes by increasing the number of truck-cargo pairings. However, these strategies fail to ensure global matching pair maximization across all regions due to the heterogeneous spatial distribution of truck supply and cargo transporting demand. This limitation necessitates the incorporation of regional supply–demand gap prediction into the matching process. Two critical challenges emerge in achieving optimal matching: (1) the prediction accuracy of supply–demand gaps is influenced by multiple complex factors, and (2) the multi-objective optimization conflicts within current matching decisions may adversely impact subsequent matching performance. To address these challenges, we propose a hierarchical reinforcement learning framework for multi-objective truck-cargo matching, comprising two key components: a supply–demand gap prediction module and a multi-objective optimization matching module.For accurate supply–demand gap prediction, we develop a hypergraph attention network model incorporating an adaptive confidence interval optimization mechanism to capture complex relationships among various predictive factors. Furthermore, to mitigate negative effects of multi-objective conflicts on long-term matching performance, we design a hierarchical deep Q network model that dynamically adjusts objective weights based on predicted long-term benefits. Extensive experiments conducted on two real-world logistics datasets demonstrate that our proposed method achieves a 10.7% higher competitive ratio compared to state-of-the-art approaches, validating the effectiveness of our supply–demand balance guided matching strategy in practical operational scenarios.

Warehouse dwell time (WDT) of a truck is a critical metric for evaluating plant-logistics efficiency, including the time of the truck's queuing outside and loading inside the warehouse. ButWDTprediction is challenging as it is affected by diverse factors like loading distinct types and weights of the cargoes, and varying amounts of loading tasks in different time slots. Besides, each trucks'WDTis transitively influenced by its preceding trucks' loading time in the queue. In this paper, we propose a multi-block dwell time prediction framework consisting ofLSTMmodel and self-attention mechanism, called SDP. In view of that low performance ofSDPbrought by sparse loading data of some warehouses, we further design a multi-source adaptation based block-to-block transfer learning module. We present a warehouse similarity measurement based on loading tasks allocated and loading ability of the warehouses, according to which we enhance overall prediction performance by learning from high-performanceWDTprediction models of similar warehouses. Experimental results on a large-scale logistics data set demonstrate that our proposal can reduce Mean Absolute Percentage Error (MAPE) by an average of 10.0%, Mean Absolute Error(MAE) by an average of 16.5%, and Root Mean Square Error(RMSE) by an average of 17.0% as compared to the baselines.

Along with the increase in strict regulation of drunk driving behavior in China, the demands for designated driving services have risen in popularity. In the absence of specialized cycling map for the designated drivers who use foldable electric bicycles, they tend to take a detour or are lost on the way to the car owners’ appointed parking places. With gradual popularization of chauffeur services, cycling trajectories generated by designated drivers almost spread all over the city. It provides a chance for inferring the cycling map dedicated to the designated drivers. However, to infer an accurate map using trajectories faces severe challenges stemming from random cycling behaviors of designated drivers, including (i) trajectories contain a lot of noises and incomplete segments, (ii) turning trajectories at minor intersections are very sparse and (iii) trajectories on the roads of distinct shapes are obviously different. To address the above challenges, we propose a three-phase map inference framework, called SAMI, consisting of trajectory refinement, intersection pinpointing, and road curve interlinking. Specifically, cycling behavioral differences from neighbor regions are incorporated into intersection identification process to ensure obtaining high detection precision even when trajectory data is sparse. Further, shape-aware based centerline fitting strategy is put forward to guarantee that inferred road curves are consistent with real road shape as possible. Finally, extensive comparative experiments on two real data sets demonstrate that SAMI significantly outperforms state-of-the-art methods by 13.31% in F1-score of map inference and by 44.88% in recall rate of minor intersection detection.

Delayed and missed detection of road closure brings a great influence on the quality of the digital map. The existing solutions using trajectory data aim to identify the closed roads according to the drastic drop property in traffic flow. But in actual applications, such methods may lead to the misidentification of a traffic jam as closure, and cannot detect some events like one side closure of two-way road and the closure in the middle of the road. With the occurrence of road closure, there are variations of turning volume of neighboring roads and the increment of U-turn frequency on the closed roads besides the drastic drop of traffic flow. In this paper, we present a high-efficiency road closure detection framework based upon multi-feature fusion, called RCDM. It consists of an off-line road closure feature modeling part and an online identification part. In the off-line phase, we first partition the road network into grids, and then extract road closure features of grids and those of roads from historical data. In the online phase, on the basis of the predictions for road closure features, we screen out closed grid candidates in terms of traffic flow plunge property and further pinpoint the closed road sections according to turning behavior variations of roads. Extensive experimental results on three real data sets from Chengdu, Shanghai and Beijing validate that our method has higher detection accuracy and efficiency compared with the existing methods.

Owing to a wide variety of deployment of GPS -enabled devices, tremendous amounts of trajectories have been generated in distributed stream manner. It opens up new opportunities to track and analyze the moving behaviors of the entities. In this work, we focus on the issue of outlier detection over distributed trajectory streams, where the outliers refer to a few entities whose motion behaviors are significantly different from their local neighbors. In view of skewed distribution property and evolving nature of trajectory data, and on-the-fly detection requirement over distributed streams, we first design a high-efficiency outlier detection solution. It consists of identifying abnormal trajectory fragment and exceptional fragment cluster at the remote sites and then detecting abnormal evolving object at the coordinator site. Further, given that outlier detection accuracy would be damaged due to using inappropriate proximity thresholds or a few trajectory data not having sufficient neighbors at the remote sites, we extract proximity thresholds of different regions and spatial context relationship of each region from historical data to improve the precision. Built upon this is an improved version consisting of off-line modeling phase and on-line detection phase. During the on-line phase, the proximity thresholds that are derived from historical trajectories during the off-line phase are leveraged to assist in detecting abnormal trajectory fragments and exceptional fragment clusters at the remote sites. Additionally, at the coordinator site, the detection results of some remote sites can be refined by incorporating those of other remote sites with neighborhood relationship. Extensive experimental results on real data demonstrate that our proposed methods own high detection validity, less communication cost and linear scalability for online identifying outliers over distributed trajectory streams.

The rapid development of steel logistics industry still has not effectively address such issues as truck overload and order overdue as well as cargo overstock. One of the reasons lie in limited number of trucks for transporting large scale cargos. More importantly, traditional methods attend to distribute cargos to trucks with the aim of maximizing the loading of each truck. But they ignore the priority level of orders and the expiration date of cargos stored in the warehouses, which have critical influences on profits of steel logistics industry. Hence, it necessitates an appropriate cargo distribution mechanism under the precondition of limited transportation capacity resources, to guarantee the maximization of delivery proportion for high-priority cargos. Recently, tremendous logistics data has been produced and are being in constant increment hourly in steel logistics platform. However, there is no existing solution to transform such data into actionable scheme to improve cargo distributing effectiveness. This paper puts forward a system implementation of smart cargo loading plan decision framework (SCLPD for short) for steel logistics industry. Through analysis on numerous real data cargo loading plan and inventory of warehouse, some important rules related to cargo distribution process are extracted. Additionally, consider that different amounts of trucks arriving in different time periods, based on adaptive time window model, a two- layer searching mechanism consisting of a genetic algorithm and A* algorithm is designed to ensure global optimization of cargo loading plan for the trucks in all time periods. In our demonstration, we illustrate the procedure of matching for cargos and trucks in various time windows, and showcase the comparison experimental results between the traditional method and SCLPD by the measurement of delivery proportion for high- priority cargos. The effectiveness and practicality of SCLPD enables efficient cargo loading plan generation, to meet the real- world requirements from steel logistics platform.

The inaccuracy of road intersection in digital road map easily brings serious effects on the mobile navigation and other applications. Massive traveling trajectories of thousands of vehicles enable frequent updating of road intersection topology. In this paper, we first expand the road intersection detection issue into a topology calibration problem for road intersection influence zone. Distinct from the existing road intersection update methods, we not only determine the location and coverage of road intersection, but figure out incorrect or missing turning paths within whole influence zone based on unmatched trajectories as compared to the existing map. The important challenges of calibration issue include that trajectories are mixing with exceptional data, and road intersections are of different sizes and shapes, etc. To address above challenges, we propose a three-phase calibration framework, called CITT. It is composed of trajectory quality improving, core zone detection, and topology calibration within road intersection influence zone. From such components it can automatically obtain high quality topology of road intersection influence zone. Extensive experiments compared with the state-of-the-art methods using trajectory data obtained from Didi Chuxing and Chicago campus shuttles demonstrate that CITT method has strong stability and robustness and significantly outperforms the existing methods.

Large collections of GPS trajectory data provide us unprecedented opportunity to detect the road intersection automatically. However, in the real-world scenarios, the precision of existing detection methods cannot be guaranteed due to severe challenges including (i) low-quality raw GPS trajectory data and (ii) the difficulty of differentiating intersections from nonintersections. To tackle above issues, we propose a novel twophase road intersection detection framework, called as RIDF, which is comprised of trajectory quality improving and intersection extracting. More importantly, through extracting candidate cells based on direction statistic analysis and refining the locations of intersections using hybrid clustering strategy, our approach can effectively detect road intersections of different size. An experimental evaluation on two real data sets extensively assesses the quality of RIDF method by comparing it with state-of-theart methods. Experimental results demonstrate that our proposal can overcome the limitations of existing methods and thus have better accuracy than the existing work.

The existing detection techniques are not tailored to identify the outlier which is close to its neighbors according to some features, but behaves significantly distinct from its neighbors in terms of the other features. In this paper, we propose a feature grouping-based mechanism, and then present two algorithms to detect outliers (TF-outlier and MO-outlier) upon trajectory streams. The effectiveness and efficiency of our proposal are validated by the experiments on real trajectory data.

The wide deployments of GPS-embedded devices have produced multiple rapid voluminous trajectory streams, which needs to be analyzed to extract abnormal behaviors of moving objects in real-time. To date, outlier detection over distributed trajectory streams has not received enough focuses due to the constraint factors like skewness distribution and evolving nature of trajectory data, and on-the-fly execution requirement with minimal communication cost. In this paper, we present the first scalable decentralized outlier detection framework over distributed trajectory streams, called ODDTS. It consists of remote site processing and coordinator processing, with the aim of continuously providing feature-grouping based outliers detection over distributed trajectory streams. Extensive experiments over real data demonstrate high detecting validity, less communication cost and linear scalability of ODDTS method for online identifying outliers upon distributed trajectory streams.

Outlier detection acts as one of the most important analysis tasks for trajectory stream. In stream scenarios, such properties as unlimitedness, time-varying evolutionary, sparsity, and skewness distribution of trajectories pose new challenges to outlier detection technique. Trajectory outlier detection techniques mainly focus on finding trajectory that is dissimilar to the majority of the others, which is based on the hypothesis that they are probably generated by a different mechanism. Most distance-based methods tend to utilize a function (e.g., weighted linear sum) to measure the similarity of two arbitrary objects provided that representative features have been extracted in advance. However, this kind of method is not tailored to identify the outlier which is close to its neighbors according to some features, but behaves significantly different from its neighbors in terms of the other features. To address this issue, we propose a feature grouping-based mechanism that divides all the features into two groups, where the first group (Similarity Feature) is used to find close neighbors and the second group (Difference Feature) is used to find outliers within the similar neighborhood. According to the feature differences among local adjacent objects in one or more time intervals, we present two outlier definitions, including local anomaly trajectory fragment (TF-outlier) and evolutionary anomaly moving object (MO-outlier ). We devise a basic solution and then an optimized algorithm to detect both types of outliers. Experimental results show that our proposal is both effective and efficient to detect outliers upon trajectory data streams.