Autonomous driving (AD) systems often exhibit driving styles that differ considerably from those of human drivers. When users engage with AD systems, they are essentially passengers in a vehicle controlled by an unseen operator. This lack of fitness can lead to unexpected driving behaviors, which may significantly undermine user trust. Evaluating human–machine fitness is essential for ensuring safety, efficiency, and user satisfaction in AD systems. However, evaluating human acceptance of AD styles is challenging due to their subjective nature and context-dependent variability. In this study, we utilize electroencephalography (EEG) signals collected from the same individuals during both autonomous and manual driving (MD) to analyze AD acceptance, capturing subtle human responses during human–machine interactions. By examining the differences between MD and AD styles in car-following segments, we quantify AD acceptance. We build a prediction model using EEG-based brain functional networks with various similarity metrics and thresholding methods. The results show that our model outperforms baseline methods, achieving an R2of 0.824, with the combination of phase-locking value and density-based thresholding being the most effective. Cognitive analysis indicates that brain networks exhibit greater efficiency during MD than AD, and this difference diminishes with increased acceptance of AD. Furthermore, high-order connectivity matrices indicate that different acceptance levels are linked to distinct coupling patterns between brain networks across various task states. This research provides insights for continuously assessing AD acceptance, aiding in the development of reliable systems tailored to diverse users.

Accurately and jointly predicting multimodal transportation demand is crucial for pre-allocating transport resources, enhancing the resilience of traffic systems. However, current approaches insufficiently explore inter- and intra-mode heterogeneity, resulting in undifferentiated dependency extraction. Moreover, existing research struggles to model cross-mode interactions among three or more transportation modes and adapt to dynamic relations in multimodal demand. To address these limitations, we propose a novel multimodal demand prediction model based on a meta-parameter learning network (MMDNet), centered on characterizing multimodal traffic spatial-temporal heterogeneity and unifying the modeling of cross-mode interactions. Our model features: 1) a spatial-temporal heterogeneity meta-parameter learning method, capturing both inter- and intra-mode heterogeneity to steer more targeted dependency extraction than previous studies; 2) a spatial-temporal evolving unified graph generator, transcending prior studies’ limitations in unifying dynamic interactions across three or more modes by creating dynamic unified graphs. Extensive experiments on three real-world datasets (New York, Beijing and Chicago) covering four different traffic modes are carried out to evaluate the MMDNet. The model achieves a 6.65% performance gain over advanced baselines and demonstrates strong cross-city adaptability. Abundant interpretability analyses show our model can semantically encode explainable cross-mode interactions and differences between modes. Source codes are available athttps://github.com/zhjiang1/MMDNet

The proliferation of the Indoor Positioning System (IPS) has enabled unprecedented opportunities for exploring individual mobility patterns (IMPs) within transport hubs. IMP prediction studies in indoor environments face three critical limitations: suffering from incompatible data cleaning, ignoring the characteristic of trajectories having fewer “stay” phases, and lacking adequate individual historical trajectories. This study proposes an IMPs prediction framework by designing a novel floor filter algorithm (FFA), a hybrid Douglas-Peucker and Density-Based Spatial Clustering of Applications with Noise (DP-DBSCAN), and a various-order Markov chain tree (VOMC-Tree) model. Within this framework, the FFA is utilized to eliminate the effects of “floor hopping”. Additionally, the DP-DBSCAN algorithm is designed to identify Key Points (KPs) within trajectories, conquering the limitation of fewer “stay” phases. To overcome the lack of individual historical trajectories, the VOMC-Tree model is proposed for online predictions of passengers’ IMPs based on offline training. To evaluate the performance of the prediction framework, a case study is conducted using a Bluetooth-based indoor trajectory dataset collected from Beijing Capital International Airport (BCIA) of China and an expanded dataset. Experimental results demonstrate that our framework outperforms baseline models across various scales of training datasets, as further evidenced by an analytical unpredictability ratio. The VOMC-Tree model performs well in predicting the sharing rate of transportation modes. The predictions of IMPs can be applied to intelligent systems for the management of transport hubs and the optimization of passenger flow. The predictions regarding the sharing rate are beneficial to the planning and scheduling of multi-traffic modes system.

Over time, significant progress has been made in planning and managing public transit networks (PTNs). However, most research developments are confined to pairwise interactions, thus offering restricted perspectives on the comprehension of higher order PTN reliability, particularly the exploration of cascading failure for multimodal PTNs (MPTNs). Hence, on the basis of the conventional coupled map lattice (CML) model, we propose a cliquey CML model in which failure propagation can occur via interactions within cliques of varying scales to investigate the higher order cascading reliability of MPTNs. In addition, models are constructed for three PTN types: bus-only, bus-metro, and bus-metro-taxi/ride-hailing networks. With Beijing MPTNs as empirical examples, we design various attack strategies to explore the resilience characteristics of MPTNs after being destroyed from different perspectives. Overall, Beijing MPTNs display favorable resilience and cascading reliability in the face of intentional attacks, and PTNs considering higher order interactions display better network stability than lower order node-line networks do because of the effect of cliques. Additionally, MPTNs generally exhibit better survivability than unimodal PTNs in scenarios with low perturbations and collapse relatively quickly in cases with high perturbations. This article provides a theoretical foundation for advancing research on the higher order dynamics of MPTNs.

Accurate prediction of demand for traditional taxi and ride-hailing services is crucial for addressing supply-demand imbalances. However, recent studies based on global adaptive graphs, local spatial-temporal graphs, and self-attention mechanisms struggle to effectively capture the dynamic and intricate relations in demand. Moreover, existing dynamic graph generators face challenges in efficiently producing high-quality graphs to learn the diverse interactions among zones along time axis and their shared patterns spanning various time scales. To solve these challenges, we propose a novel Local-Perception-Enhanced Spatial-Temporal Evolving Graph Transformer Network (LPE-STGTN), aimed at improving the effectiveness and efficiency of extracting intricate local dependencies in taxi demand. Specifically, we elaborately design a spatial-temporal evolving graph generator to absorb shared and diversified inter-zone relations across different temporal periodicities and specific interactions among zones within each time step. Furthermore, an attention free transformer with local context (AFT-local) is introduced to effectively learn the correlations between adjacent time steps. Extensive experiments on three taxi datasets of New York and Beijing are carried out to evaluate the superior performance of our model. Compared with the most competitive baseline, our model achieves a balance between effectiveness and efficiency on three datasets, with average training time reduction of 70.66% and average performance improvement of 1.96%.

The ubiquity of mobile phone signaling data (MPSD) allows us to study travel mode identification (TMI) of a larger scale of population in cities than GPS data and travel survey. Existing studies suffers from deficient data cleaning, ignoring low spatial accuracy of data and lack of the common datasets labeled with travel modes. This study proposes a novel TMI framework with MPSD and pseudo MPSD by designing hidden markov model (HMM)-based map-matching algorithm and bi-directional long short-term memory (Bi-LSTM)-based deep neural network (DNN). In this framework, conventional data cleaning method is modified to eliminate outliers in raw MPSD to improve data quality. In addition, HMM-based map-matching algorithm is designed to estimate the actual position of travelers from MPSD for spatial-temporal accuracy improvement. Specifically, pseudo MPSD is reconstructed by multi-source data aiming at overcoming the facts lacking of labeled MPSD. Both MPSD and labeled pseudo MPSD are processed to features, where the latter is used to train model based on Bi-LSTM, and the former is input to the trained model for TMI application validation. To evaluate the framework performance, a case study is carried out. Experimental results show that our framework outperforms baseline models on the same datasets. Furthermore, the accuracy improvement of designed approaches is demonstrated by a series of algorithm comparisons. The framework has good generality and flexibility, and can be extended to other cities for TMI.

Knowledge about mobility patterns has become increasingly important to urban development. In this work, public transit origin-destination (OD) mobility patterns are undergoing meso-level analysis in using the advantages of big data and for the creation of a new planning and decision-based tool. An ensemble clustering method is proposed to abstract the common OD pairs by fully considering link-based information, and the nonnegative tensor factorization model is adopted to effectively extract and visualize quantitatively the mobility patterns of OD pairs. This is attained by using multi-layered analysis such as of traffic demand, traffic accessibility and traffic congestion to enable different visual and quantitative mobility patterns. In the case study of Beijing, these patterns were analyzed and discussed by temporal and spatial factors. The results of the various patterns show explicitly when and where to provide remedies to traffic problems, by time and space. This is analogue, to some extent, to detecting and treating black spots of road accidents. The new developed multi-layered mesoscopic analysis could, therefore, be an important tool for improving urban planning, public transit planning, traffic management, and emergency intervention.

Mobility patterns at region level can provide more macroscopic and intuitive knowledge on how people gather in or depart from the region. However, the analysis and prediction of regional mobility patterns have yet to be effectively addressed. In light of this, using smart card data (SCD) and points of interest (POI) data, a multi-step methodology which integrates the inner-restricted fuzzy C-means clustering, nonnegative tensor factorization and artificial neural network are proposed and implemented in this paper. It overcomes the difficulties in region division, pattern extraction, and prediction. The bus SCD and POI data in Beijing city are utilized for proving the usefulness of the methodology. The regional mobility patterns of bus travellers in Beijing city are extracted from the third-order tensors involving 1110 regions, 34 time slots, and 7 days of the week. The analyzed results show that the proposed methodology has a good performance on predicting the regional mobility patterns based on the regional properties. Furthermore, by considering both of the regional boarding and alighting patterns, the predictions of the regional aggregation pattern can also be achieved. These research achievements can not only provide a deep insight on the human mobility patterns at region level, but also support the evidence-based and forward-looking urban planning and intelligent transportation management.

In this paper, many similar freight demand in urban area are firstly collaborated and mixed in terms of freight classification rules.And then, with the consideration of vehicle utilization and customer logistics cost, the paper proposes a city freight intelligent scheduling model which simultaneously solve two problems of freight load and path optimization.Each customer requires the freights to be picked up and delivered by the same vehicle within a given time window.At last, the model is solved by Genetic Algorithm.A study case is introduced to show the results.The results provide a set of distribution scheme including load plan, pickup plan and delivery plan.