arXiv subjects

Lijun Sun

Publications and source records attributed to Lijun Sun.

At least 91 records · Page 5Linked to original sources

arXiv2022

Probabilistic forecasting of bus travel time with a Bayesian Gaussian mixture model

Accurate forecasting of bus travel time and its uncertainty is critical to service quality and operation of transit systems; for example, it can help passengers make better decisions on departure time, route choice, and even transport mode choice and also support transit operators to make informed decisions on tasks such as crew/vehicle scheduling and timetabling. However, most existing approaches in bus travel time forecasting are based on deterministic models that provide only point estimation. To this end, we develop in this paper a Bayesian probabilistic forecasting model for bus travel time. To characterize the strong dependencies/interactions between consecutive buses, we concatenate the link travel time vectors and the headway vector from a pair of two adjacent buses as a new augmented variable and model it with a constrained Multivariate Gaussian mixture distributions. This approach can naturally capture the interactions between adjacent buses (e.g., correlated speed and smooth variation of headway), handle missing values in data, and depict the multimodality in bus travel time distributions. Next, we assume different periods in a day share the same set of Gaussian components but different mixing coefficients to characterize the systematic temporal variations in bus operation. For model inference, we develop an efficient Markov chain Monte Carlo (MCMC) sampling algorithm to obtain the posterior distributions of model parameters and make probabilistic forecasting. We test the proposed model using the data from a twenty-link bus route in Guangzhou, China. Results show our approach significantly outperforms baseline models that overlook bus-to-bus interactions in terms of both predictive means and distributions. Besides forecasting, the parameters of the proposed model contain rich information for understanding/improving the bus service.

stat.AP↗

arXiv2022

Real-time forecasting of metro origin-destination matrices with high-order weighted dynamic mode decomposition

Forecasting the short-term ridership among origin-destination pairs (OD matrix) of a metro system is crucial in real-time metro operation. However, this problem is notoriously difficult due to the high-dimensional, sparse, noisy, and skewed nature of OD matrices. This paper proposes a High-order Weighted Dynamic Mode Decomposition (HW-DMD) model for short-term metro OD matrices forecasting. DMD uses Singular Value Decomposition (SVD) to extract low-rank approximation from OD data, and a low-rank high-order vector autoregression model is established for forecasting. To address a practical issue that metro OD matrices cannot be observed in real-time, we use the boarding demand to replace the unavailable OD matrices. Particularly, we consider the time-evolving feature of metro systems and improve the forecast by exponentially reducing the weights for old data. Moreover, we develop a tailored online update algorithm for HW-DMD to update the model coefficients daily without storing historical data or retraining. Experiments on data from a large-scale metro system show the proposed HW-DMD is robust to the noisy and sparse data and significantly outperforms baseline models in forecasting both OD matrices and boarding flow. The online update algorithm also shows consistent accuracy over a long time when maintaining an HW-DMD model at low costs.

stat.AP↗

arXiv2022

Bayesian inference for link travel time correlation of a bus route

Estimation of link travel time correlation of a bus route is essential to many bus operation applications, such as timetable scheduling, travel time forecasting and transit service assessment/improvement. Most previous studies rely on either independent assumptions or simplified local spatial correlation structures. In the real world, however, link travel time on a bus route could exhibit complex correlation structures, such as long-range correlations, negative correlations, and time-varying correlations. Therefore, before introducing strong assumptions, it is essential to empirically quantify and examine the correlation structure of link travel time from real-world bus operation data. To this end, this paper develops a Bayesian Gaussian model to estimate the link travel time correlation matrix of a bus route using smart-card-like data. Our method overcomes the small-sample-size problem in correlation matrix estimation by borrowing/integrating those incomplete observations (i.e., with missing/ragged values and overlapped link segments) from other bus routes. Next, we propose an efficient Gibbs sampling framework to marginalize over the missing and ragged values and obtain the posterior distribution of the correlation matrix. Three numerical experiments are conducted to evaluate model performance. We first conduct a synthetic experiment and our results show that the proposed method produces an accurate estimation for travel time correlations with credible intervals. Next, we perform experiments on a real-world bus route with smart card data; our results show that both local and long-range correlations exist on this bus route. Finally, we demonstrate an application of using the estimated covariance matrix to make probabilistic forecasting of link and trip travel time.

stat.AP↗

arXiv2021

Robust Dynamic Bus Control: A Distributional Multi-agent Reinforcement Learning Approach

Bus system is a critical component of sustainable urban transportation. However, the operation of a bus fleet is unstable in nature, and bus bunching has become a common phenomenon that undermines the efficiency and reliability of bus systems. Recently research has demonstrated the promising application of multi-agent reinforcement learning (MARL) to achieve efficient vehicle holding control to avoid bus bunching. However, existing studies essentially overlook the robustness issue resulting from various events, perturbations and anomalies in a transit system, which is of utmost importance when transferring the models for real-world deployment/application. In this study, we integrate implicit quantile network and meta-learning to develop a distributional MARL framework -- IQNC-M -- to learn continuous control. The proposed IQNC-M framework achieves efficient and reliable control decisions through better handling various uncertainties/events in real-time transit operations. Specifically, we introduce an interpretable meta-learning module to incorporate global information into the distributional MARL framework, which is an effective solution to circumvent the credit assignment issue in the transit system. In addition, we design a specific learning procedure to train each agent within the framework to pursue a robust control policy. We develop simulation environments based on real-world bus services and passenger demand data and evaluate the proposed framework against both traditional holding control models and state-of-the-art MARL models. Our results show that the proposed IQNC-M framework can effectively handle the various extreme events, such as traffic state perturbations, service interruptions, and demand surges, thus improving both efficiency and reliability of the system.

cs.LG↗

arXiv2021

Efficient Connected and Automated Driving System with Multi-agent Graph Reinforcement Learning

Connected and automated vehicles (CAVs) have attracted more and more attention recently. The fast actuation time allows them having the potential to promote the efficiency and safety of the whole transportation system. Due to technical challenges, there will be a proportion of vehicles that can be equipped with automation while other vehicles are without automation. Instead of learning a reliable behavior for ego automated vehicle, we focus on how to improve the outcomes of the total transportation system by allowing each automated vehicle to learn cooperation with each other and regulate human-driven traffic flow. One of state of the art method is using reinforcement learning to learn intelligent decision making policy. However, direct reinforcement learning framework cannot improve the performance of the whole system. In this article, we demonstrate that considering the problem in multi-agent setting with shared policy can help achieve better system performance than non-shared policy in single-agent setting. Furthermore, we find that utilization of attention mechanism on interaction features can capture the interplay between each agent in order to boost cooperation. To the best of our knowledge, while previous automated driving studies mainly focus on enhancing individual's driving performance, this work serves as a starting point for research on system-level multi-agent cooperation performance using graph information sharing. We conduct extensive experiments in car-following and unsignalized intersection settings. The results demonstrate that CAVs controlled by our method can achieve the best performance against several state of the art baselines.

stat.ML↗

arXiv2021

Hankel-structured Tensor Robust PCA for Multivariate Traffic Time Series Anomaly Detection

Spatiotemporal traffic data (e.g., link speed/flow) collected from sensor networks can be organized as multivariate time series with additional spatial attributes. A crucial task in analyzing such data is to identify and detect anomalous observations and events from the data with complex spatial and temporal dependencies. Robust Principal Component Analysis (RPCA) is a widely used tool for anomaly detection. However, the traditional RPCA purely relies on the global low-rank assumption while ignoring the local temporal correlations. In light of this, this study proposes a Hankel-structured tensor version of RPCA for anomaly detection in spatiotemporal data. We treat the raw data with anomalies as a multivariate time series matrix (location $\times$ time) and assume the denoised matrix has a low-rank structure. Then we transform the low-rank matrix to a third-order tensor by applying temporal Hankelization. In the end, we decompose the corrupted matrix into a low-rank Hankel tensor and a sparse matrix. With the Hankelization operation, the model can simultaneously capture the global and local spatiotemporal correlations and exhibit more robust performance. We formulate the problem as an optimization problem and use tensor nuclear norm (TNN) to approximate the tensor rank and $l_1$ norm to approximate the sparsity. We develop an efficient solution algorithm based on the Alternating Direction Method of Multipliers (ADMM). Despite having three hyper-parameters, the model is easy to set in practice. We evaluate the proposed method by synthetic data and metro passenger flow time series and the results demonstrate the accuracy of anomaly detection.

cs.LG↗

arXiv2021

Spatial Aggregation and Temporal Convolution Networks for Real-time Kriging

Spatiotemporal kriging is an important application in spatiotemporal data analysis, aiming to recover/interpolate signals for unsampled/unobserved locations based on observed signals. The principle challenge for spatiotemporal kriging is how to effectively model and leverage the spatiotemporal dependencies within the data. Recently, graph neural networks (GNNs) have shown great promise for spatiotemporal kriging tasks. However, standard GNNs often require a carefully designed adjacency matrix and specific aggregation functions, which are inflexible for general applications/problems. To address this issue, we present SATCN -- Spatial Aggregation and Temporal Convolution Networks -- a universal and flexible framework to perform spatiotemporal kriging for various spatiotemporal datasets without the need for model specification. Specifically, we propose a novel spatial aggregation network (SAN) inspired by Principal Neighborhood Aggregation, which uses multiple aggregation functions to help one node gather diverse information from its neighbors. To exclude information from unsampled nodes, a masking strategy that prevents the unsampled sensors from sending messages to their neighborhood is introduced to SAN. We capture temporal dependencies by the temporal convolutional networks, which allows our model to cope with data of diverse sizes. To make SATCN generalizable to unseen nodes and even unseen graph structures, we employ an inductive strategy to train SATCN. We conduct extensive experiments on three real-world spatiotemporal datasets, including traffic speed and climate recordings. Our results demonstrate the superiority of SATCN over traditional and GNN-based kriging models.

cs.LG↗

arXiv2021

Dynamic Spatiotemporal Graph Convolutional Neural Networks for Traffic Data Imputation with Complex Missing Patterns

Missing data is an inevitable and ubiquitous problem for traffic data collection in intelligent transportation systems. Despite extensive research regarding traffic data imputation, there still exist two limitations to be addressed: first, existing approaches fail to capture the complex spatiotemporal dependencies in traffic data, especially the dynamic spatial dependencies evolving with time; second, prior studies mainly focus on randomly missing patterns while other more complex missing scenarios are less discussed. To fill these research gaps, we propose a novel deep learning framework called Dynamic Spatiotemporal Graph Convolutional Neural Networks (DSTGCN) to impute missing traffic data. The model combines the recurrent architecture with graph-based convolutions to model the spatiotemporal dependencies. Moreover, we introduce a graph structure estimation technique to model the dynamic spatial dependencies from real-time traffic information and road network structure. Extensive experiments based on two public traffic speed datasets are conducted to compare our proposed model with state-of-the-art deep learning approaches in four types of missing patterns. The results show that our proposed model outperforms existing deep learning models in all kinds of missing scenarios and the graph structure estimation technique contributes to the model performance. We further compare our proposed model with a tensor factorization model and find distinct behaviors across different model families under different training schemes and data availability.

cs.LG↗

arXiv2021

Development of water extraction system for liquid scintillatorpurification of JUNO

The Jiangmen Underground Neutrino Observatory (JUNO) uses 20k tons of liquid scintillator (LS)to detect neutrinos. The content of radioactive substances in the liquid scintillator will affect theexperimental results. JUNO will use counter current water extraction to reduce the radioactive metalions inside the LS. In this article, The factors that affect the final water extraction like the partitioncoefficient and the factors that affect it, the optimal mass transfer droplet size, the flow rate ratio,theoretical stage and different working mode (LS as continuous phase or dispersed phase) have beenstudied. We built counter current extraction prototype in the laboratory and a pilot plant in Daya Bay.We not only study the factors above, but also get a lot of engineering experience that worth sharing.

physics.ins-det↗

arXiv2021

Spatially Focused Attack against Spatiotemporal Graph Neural Networks

Spatiotemporal forecasting plays an essential role in various applications in intelligent transportation systems (ITS), such as route planning, navigation, and traffic control and management. Deep Spatiotemporal graph neural networks (GNNs), which capture both spatial and temporal patterns, have achieved great success in traffic forecasting applications. Understanding how GNNs-based forecasting work and the vulnerability and robustness of these models becomes critical to real-world applications. For example, if spatiotemporal GNNs are vulnerable in real-world traffic prediction applications, a hacker can easily manipulate the results and cause serious traffic congestion and even a city-scale breakdown. However, despite that recent studies have demonstrated that deep neural networks (DNNs) are vulnerable to carefully designed perturbations in multiple domains like objection classification and graph representation, current adversarial works cannot be directly applied to spatiotemporal forecasting due to the causal nature and spatiotemporal mechanisms in forecasting models. To fill this gap, in this paper we design Spatially Focused Attack (SFA) to break spatiotemporal GNNs by attacking a single vertex. To achieve this, we first propose the inverse estimation to address the causality issue; then, we apply genetic algorithms with a universal attack method as the evaluation function to locate the weakest vertex; finally, perturbations are generated by solving an inverse estimation-based optimization problem. We conduct experiments on real-world traffic data and our results show that perturbations in one vertex designed by SA can be diffused into a large part of the graph.

cs.LG↗

arXiv2021

Individual Mobility Prediction via Attentive Marked Temporal Point Processes

Individual mobility prediction is an essential task for transportation demand management and traffic system operation. There exist a large body of works on modeling location sequence and predicting the next location of users; however, little attention is paid to the prediction of the next trip, which is governed by the strong spatiotemporal dependencies between diverse attributes, including trip start time $t$, origin $o$, and destination $d$. To fill this gap, in this paper we propose a novel point process-based model -- Attentive Marked temporal point processes (AMTPP) -- to model human mobility and predict the whole trip $(t,o,d)$ in a joint manner. To encode the influence of history trips, AMTPP employs the self-attention mechanism with a carefully designed positional embedding to capture the daily/weekly periodicity and regularity in individual travel behavior. Given the unique peaked nature of inter-event time in human behavior, we use an asymmetric log-Laplace mixture distribution to precisely model the distribution of trip start time $t$. Furthermore, an origin-destination (OD) matrix learning block is developed to model the relationship between every origin and destination pair. Experimental results on two large metro trip datasets demonstrate the superior performance of AMTPP.

cs.LG↗

arXiv2021

Cooperative coevolution of real predator robots and virtual robots in the pursuit domain

The pursuit domain, or predator-prey problem is a standard testbed for the study of coordination techniques. In spite that its problem setup is apparently simple, it is challenging for the research of the emerged swarm intelligence. This paper presents a particle swarm optimization (PSO) based cooperative coevolutionary algorithm for the (predator) robots, called CCPSO-R, where real and virtual robots coexist in an evolutionary algorithm (EA). Virtual robots sample and explore the vicinity of the corresponding real robots and act as their action spaces, while the real robots consist of the real predators who actually pursue the prey robot without fixed behavior rules under the immediate guidance of the fitness function, which is designed in a modular manner with very limited domain knowledge. In addition, kinematic limits and collision avoidance considerations are integrated into the update rules of robots. Experiments are conducted on a scalable swarm of predator robots with 4 types of preys, the results of which show the reliability, generality, and scalability of the proposed CCPSO-R. Comparison with a representative dynamic path planning based algorithm Multi-Agent Real-Time Pursuit (MAPS) further shows the effectiveness of CCPSO-R. Finally, the codes of this paper are public available at: https://github.com/LijunSun90/pursuitCCPSOR.

cs.RO↗

arXiv2021

On Social Interactions of Merging Behaviors at Highway On-Ramps in Congested Traffic

Merging at highway on-ramps while interacting with other human-driven vehicles is challenging for autonomous vehicles (AVs). An efficient route to this challenge requires exploring and exploiting knowledge of the interaction process from demonstrations by humans. However, it is unclear what information (or environmental states) is utilized by the human driver to guide their behavior throughout the whole merging process. This paper provides quantitative analysis and evaluation of the merging behavior at highway on-ramps with congested traffic in a volume of time and space. Two types of social interaction scenarios are considered based on the social preferences of surrounding vehicles: courteous and rude. The significant levels of environmental states for characterizing the interactive merging process are empirically analyzed based on the real-world INTERACTION dataset. Experimental results reveal two fundamental mechanisms in the merging process: 1) Human drivers select different states to make sequential decisions at different moments of task execution, and 2) the social preference of surrounding vehicles can impact variable selection for making decisions. It implies that efficient decision-making design should filter out irrelevant information while considering social preference to achieve comparable human-level performance. These essential findings shed light on developing new decision-making approaches for AVs.

cs.RO↗

arXiv2021

Scalable Low-Rank Tensor Learning for Spatiotemporal Traffic Data Imputation

Missing value problem in spatiotemporal traffic data has long been a challenging topic, in particular for large-scale and high-dimensional data with complex missing mechanisms and diverse degrees of missingness. Recent studies based on tensor nuclear norm have demonstrated the superiority of tensor learning in imputation tasks by effectively characterizing the complex correlations/dependencies in spatiotemporal data. However, despite the promising results, these approaches do not scale well to large data tensors. In this paper, we focus on addressing the missing data imputation problem for large-scale spatiotemporal traffic data. To achieve both high accuracy and efficiency, we develop a scalable tensor learning model -- Low-Tubal-Rank Smoothing Tensor Completion (LSTC-Tubal) -- based on the existing framework of Low-Rank Tensor Completion, which is well-suited for spatiotemporal traffic data that is characterized by multidimensional structure of location$\times$ time of day $\times$ day. In particular, the proposed LSTC-Tubal model involves a scalable tensor nuclear norm minimization scheme by integrating linear unitary transformation. Therefore, tensor nuclear norm minimization can be solved by singular value thresholding on the transformed matrix of each day while the day-to-day correlation can be effectively preserved by the unitary transform matrix. We compare LSTC-Tubal with state-of-the-art baseline models, and find that LSTC-Tubal can achieve competitive accuracy with a significantly lower computational cost. In addition, the LSTC-Tubal will also benefit other tasks in modeling large-scale spatiotemporal traffic data, such as network-level traffic forecasting.

stat.ML↗

arXiv2021

Incorporating travel behavior regularity into passenger flow forecasting

Accurate forecasting of passenger flow (i.e., ridership) is critical to the operation of urban metro systems. Previous studies mainly model passenger flow as time series by aggregating individual trips and then perform forecasting based on the values in the past several steps. However, this approach essentially overlooks the fact that passenger flow consists of trips from each individual traveler. For example, a traveler's work trip in the morning can help predict his/her home trip in the evening, while this causal structure cannot be explicitly encoded in standard time series models. In this paper, we propose a new forecasting framework for boarding flow by incorporating the generative mechanism into standard time series models and leveraging the strong regularity rooted in travel behavior. In doing so, we introduce returning flow from previous alighting trips as a new covariate, which captures the causal structure and long-range dependencies in passenger flow data based on travel behavior. We develop the return probability parallelogram (RPP) to summarize the causal relationships and estimate the return flow. The proposed framework is evaluated using real-world passenger flow data, and the results confirm that the returning flow -- a single covariate -- can substantially and consistently improve various forecasting tasks, including one-step ahead forecasting, multi-step ahead forecasting, and forecasting under special events. And the proposed method is more effective for business-type stations with most passengers come and return within the same day. This study can be extended to other modes of transport, and it also sheds new light on general demand time series forecasting problems, in which causal structure and long-range dependencies are generated by the user behavior.

stat.AP↗

arXiv2021

Quantifying out-of-station waiting time in oversaturated urban metro systems

Metro systems in megacities such as Beijing, Shenzhen and Guangzhou are under great passenger demand pressure. During peak hours, it is common to see oversaturated conditions (i.e., passenger demand exceeds network capacity), which bring significant operational risks and safety issues. A popular control intervention is to restrict the entering rate during peak hours by setting up out-of-station queueing with crowd control barriers. The \textit{out-of-station waiting} can make up a substantial proportion of total travel time but is not well-studied in the literature. Accurate quantification of out-of-station waiting time is important to evaluating the social benefit and cost of service scheduling/optimization plans; however, out-of-station waiting time is difficult to estimate because it is not a part of smart card transactions. In this study, we propose an innovative method to estimate the out-of-station waiting time by leveraging the information from a small group of transfer passengers -- those who transfer from nearby bus routes to the metro station. Based on the estimated transfer time for this small group, we first infer the out-of-station waiting time for all passengers by developing a Gaussian Process regression with a Student-$t$ likelihood and then use the estimated out-of-station waiting time to build queueing diagrams. We apply our method to the Tiantongyuan North station of Beijing metro as a case study; our results show that the maximum out-of-station waiting time can reach 15 minutes, and the maximum queue length can be over 3000 passengers. Our results suggest that out-of-station waiting can cause significant travel costs and thus should be considered in analyzing transit performance, mode choice, and social benefits. To the best of our knowledge, this paper is the first quantitative study for out-of-station waiting time.

stat.AP↗

arXiv2021

Reducing Bus Bunching with Asynchronous Multi-Agent Reinforcement Learning

The bus system is a critical component of sustainable urban transportation. However, due to the significant uncertainties in passenger demand and traffic conditions, bus operation is unstable in nature and bus bunching has become a common phenomenon that undermines the reliability and efficiency of bus services. Despite recent advances in multi-agent reinforcement learning (MARL) on traffic control, little research has focused on bus fleet control due to the tricky asynchronous characteristic -- control actions only happen when a bus arrives at a bus stop and thus agents do not act simultaneously. In this study, we formulate route-level bus fleet control as an asynchronous multi-agent reinforcement learning (ASMR) problem and extend the classical actor-critic architecture to handle the asynchronous issue. Specifically, we design a novel critic network to effectively approximate the marginal contribution for other agents, in which graph attention neural network is used to conduct inductive learning for policy evaluation. The critic structure also helps the ego agent optimize its policy more efficiently. We evaluate the proposed framework on real-world bus services and actual passenger demand derived from smart card data. Our results show that the proposed model outperforms both traditional headway-based control methods and existing MARL methods.

cs.LG↗

arXiv2021

Low-Rank Autoregressive Tensor Completion for Spatiotemporal Traffic Data Imputation

Spatiotemporal traffic time series (e.g., traffic volume/speed) collected from sensing systems are often incomplete with considerable corruption and large amounts of missing values, preventing users from harnessing the full power of the data. Missing data imputation has been a long-standing research topic and critical application for real-world intelligent transportation systems. A widely applied imputation method is low-rank matrix/tensor completion; however, the low-rank assumption only preserves the global structure while ignores the strong local consistency in spatiotemporal data. In this paper, we propose a low-rank autoregressive tensor completion (LATC) framework by introducing \textit{temporal variation} as a new regularization term into the completion of a third-order (sensor $\times$ time of day $\times$ day) tensor. The third-order tensor structure allows us to better capture the global consistency of traffic data, such as the inherent seasonality and day-to-day similarity. To achieve local consistency, we design the temporal variation by imposing an AR($p$) model for each time series with coefficients as learnable parameters. Different from previous spatial and temporal regularization schemes, the minimization of temporal variation can better characterize temporal generative mechanisms beyond local smoothness, allowing us to deal with more challenging scenarios such "blackout" missing. To solve the optimization problem in LATC, we introduce an alternating minimization scheme that estimates the low-rank tensor and autoregressive coefficients iteratively. We conduct extensive numerical experiments on several real-world traffic data sets, and our results demonstrate the effectiveness of LATC in diverse missing scenarios.

cs.LG↗