Deep Siamese Metric Learning: A Highly Scalable Approach to Searching Unordered Sets of Trajectories

This work proposes metric learning for fast similarity-based scene retrieval of unstructured ensembles of trajectory data from large databases. We present a novel representation learning approach using Siamese Metric Learning that approximates a distance preserving low-dimensional representation and that learns to estimate reasonable solutions to the assignment problem. To this end, we employ a Temporal Convolutional Network architecture that we extend with a gating mechanism to enable learning from sparse data, leading to solutions to the assignment problem exhibiting varying degrees of sparsity. Our experimental results on professional soccer tracking data provides insights on learned features and embeddings, as well as on generalization, sensitivity, and network architectural considerations. Our low approximation errors for learned representations and the interactive performance with retrieval times several magnitudes smaller shows that we outperform previous state of the art.

Let Trajectories Speak Out the Traffic Bottlenecks

Traffic bottlenecks are a set of road segments that have an unacceptable level of traffic caused by a poor balance between road capacity and traffic volume. A huge volume of trajectory data which captures realtime traffic conditions in road networks provides promising new opportunities to identify the traffic bottlenecks. In this paper, we define this problem as trajectory-driven traffic bottleneck identification : Given a road network R , a trajectory database T , find a representative set of seed edges of size K of traffic bottlenecks that influence the highest number of road segments not in the seed set. We show that this problem is NP-hard and propose a framework to find the traffic bottlenecks as follows. First, a traffic spread model is defined which represents changes in traffic volume for each road segment over time. Then, the traffic diffusion probability between two connected segments and the residual ratio of traffic volume for each segment can be computed using historical trajectory data. We then propose two different algorithmic approaches to solve the problem. The first one is a best-first algorithm BF , with an approximation ratio of 1-1/ e . To further accelerate the identification process in larger datasets, we also propose a sampling-based greedy algorithm SG . Finally, comprehensive experiments using three different datasets compare and contrast various solutions, and provide insights into important efficiency and effectiveness trade-offs among the respective methods.

Origin-Aware Location Prediction Based on Historical Vehicle Trajectories

Next location prediction is of great importance for many location-based applications and provides essential intelligence to various businesses. In previous studies, a common approach to next location prediction is to learn the sequential transitions with massive historical trajectories based on conditional probability. Nevertheless, due to the time and space complexity, these methods (e.g., Markov models) only utilize the just passed locations to predict next locations, neglecting earlier passed locations in the trajectory. In this work, we seek to enhance the prediction performance by incorporating the travel time from all the passed locations in the query trajectory to each candidate next location. To this end, we propose a novel prediction method, namely the Travel Time Difference Model, which exploits the difference between the shortest travel time and the actual travel time to predict next locations. Moreover, we integrate the Travel Time Difference Model with a Sequential and Temporal Predictor to yield a joint model. The joint prediction model integrates local sequential transitions, temporal regularity, and global travel time information in the trajectory for the next location prediction problem. We have conducted extensive experiments on two real-world datasets: the vehicle passage record data and the taxi trajectory data. The experimental results demonstrate significant improvements in prediction accuracy over baseline methods.

PCT-TEE: Trajectory-based Private Contact Tracing System with Trusted Execution Environment

Existing Bluetooth-based private contact tracing (PCT) systems can privately detect whether people have come into direct contact with patients with COVID-19. However, we find that the existing systems lack functionality and flexibility , which may hurt the success of contact tracing. Specifically, they cannot detect indirect contact (e.g., people may be exposed to COVID-19 by using a contaminated sheet at a restaurant without making direct contact with the infected individual); they also cannot flexibly change the rules of “risky contact,” such as the duration of exposure or the distance (both spatially and temporally) from a patient with COVID-19 that is considered to result in a risk of exposure, which may vary with the environmental situation. In this article, we propose an efficient and secure contact tracing system that enables us to trace both direct contact and indirect contact. To address the above problems, we need to utilize users’ trajectory data for PCT, which we call trajectory-based PCT . We formalize this problem as a spatiotemporal private set intersection that satisfies both the security and efficiency requirements. By analyzing different approaches such as homomorphic encryption, which could be extended to solve this problem, we identify the trusted execution environment (TEE) as a candidate method to achieve our requirements. The major challenge is how to design algorithms for a spatiotemporal private set intersection under the limited secure memory of the TEE. To this end, we design a TEE-based system with flexible trajectory data encoding algorithms. Our experiments on real-world data show that the proposed system can process hundreds of queries on tens of millions of records of trajectory data within a few seconds.

A Trajectory Evaluator by Sub-tracks for Detecting VOT-based Anomalous Trajectory

With the popularization of visual object tracking (VOT), more and more trajectory data are obtained and have begun to gain widespread attention in the fields of mobile robots, intelligent video surveillance, and the like. How to clean the anomalous trajectories hidden in the massive data has become one of the research hotspots. Anomalous trajectories should be detected and cleaned before the trajectory data can be effectively used. In this article, a Trajectory Evaluator by Sub-tracks (TES) for detecting VOT-based anomalous trajectory is proposed. Feature of Anomalousness is defined and described as the Eigenvector of classifier to filter Track Lets anomalous trajectory and IDentity Switch anomalous trajectory, which includes Feature of Anomalous Pose and Feature of Anomalous Sub-tracks (FAS). In the comparative experiments, TES achieves better results on different scenes than state-of-the-art methods. Moreover, FAS makes better performance than point flow, least square method fitting and Chebyshev Polynomial Fitting. It is verified that TES is more accurate and effective and is conducive to the sub-tracks trajectory data analysis.

RealTE: Real-time Trajectory Estimation Over Large-Scale Road Networks

Real-time traffic estimation focuses on predicting the travel time of one travel path, which is capable of helping drivers selecting an appropriate or favor path. Statistical analysis or neural network approaches have been explored to predict the travel time on a massive volume of traffic data. These methods need to be updated when the traffic varies frequently, which incurs tremendous overhead. We build a system RealTER⁢e⁢a⁢l⁢T⁢E, implemented on a popular and open source streaming system StormS⁢t⁢o⁢r⁢m to quickly deal with high speed trajectory data. In RealTER⁢e⁢a⁢l⁢T⁢E, we propose a locality-sensitive partition and deployment algorithm for a large road network. A histogram estimation approach is adopted to predict the traffic. This approach is general and able to be incremental updated in parallel. Extensive experiments are conducted on six real road networks and the results illustrate RealTE achieves higher throughput and lower prediction error than existing methods. The runtime of a traffic estimation is less than 11 seconds over a large road network and it takes only 619619 microseconds for model updates.

Effects of traffic signal coordination on the safety performance of urban arterials

Urban traffic congestion and crashes have been considered by city planners as critical challenges to the economic development of the city. Traffic signal coordination, which connects a series of signals along an arterial by various coordination methodologies, has been proved as one of the most cost-effective means of reducing traffic congestion. In this regard, Metropolitan Planning Organizations (MPO) or Transportation Management Centers (TMC) have included signal timing coordination in their strategic plans. Nevertheless, concerns on the safety effects of traffic signal coordination have been continuously raised by both transportation agencies and the public. This is mainly because signal coordination may increase the travel speed along an arterial, which increases the risk and severity of traffic collisions. To date, there is neither solid evidence from the field to support the concern, nor theoretical-level models to analyze this issue. This research aims to investigate the effects of traffic signal coordination on the safety performance of urban arterials through microsimulation modeling of two traffic operational conditions: free signal operation and coordinated signals, respectively. Three urban arterials in Reno, Nevada were selected as the simulation testbed and were coded in the PTV VISSIM software. The simulated trajectory data were analyzed by the Surrogate Safety Assessment Model (SSAM) to estimate the number of traffic conflicts. Sensitivity analyses were conducted for various traffic demand levels. Results show that under unsaturated conditions, traffic signal coordination could reduce the number of conflicts in comparison with the free signal operation condition. However, under oversaturated conditions, no significant difference was found between coordinated and free signal operations. Findings from this research indicate that traffic signal coordination has the potential to reduce the risk of crashes on urban arterials under unsaturated conditions.

Data Mining Algorithm for Moving Object Trajectory

Aiming at the poor effect and long recognition time of data mining algorithm for moving target trajectory recognition, a data mining algorithm based on improved Hausdorff distance is proposed. The position and angle of abnormal trajectory data are detected by calculating the distance between trajectory classification and sub trajectory line segments, and the trajectory unit is established by using the improved Hausdorff distance algorithm to optimize the similarity matching structure. Experimental results show that the algorithm has low error pruning rate in identifying moving target trajectory, improves the detection efficiency of moving target trajectory recognition data, and ensures the quality of moving target trajectory recognition data mining

Modelling imperfect knowledge via location semantics for realistic privacy risks estimation in trajectory data

Mobility patterns of vehicles and people provide powerful data sources for location-based services such as fleet optimization and traffic flow analysis. Location-based service providers must balance the value they extract from trajectory data with protecting the privacy of the individuals behind those trajectories. Reaching this goal requires measuring accurately the values of utility and privacy. Current measurement approaches assume adversaries with perfect knowledge, thus overestimate the privacy risk. To address this issue, we introduce a model of an adversary with imperfect knowledge about the target. The model is based on equivalence areas, spatio-temporal regions with a semantic meaning, e.g. the target’s home, whose size and accuracy determine the skill of the adversary. We then derive the standard privacy metrics of k-anonymity, l-diversity and t-closeness from the definition of equivalence areas. These metrics can be computed on any dataset, irrespective of whether and what kind of anonymization has been applied to it. This work is of high relevance to all service providers acting as processors of trajectory data who want to manage privacy risks and optimize the privacy vs. utility trade-off of their services.