A “semi-lazy” approach to probabilistic path prediction in dynamic environments

2015

254

137

Data-driven intelligent transportation systems utilize data resources generated within intelligent systems to improve the performance of transportation systems and provide convenient and reliable services. Traffic data refer to datasets generated and collected on moving vehicles and objects. Data visualization is an efficient means to represent distributions and structures of datasets and reveal hidden patterns in the data. This paper introduces the basic concept and pipeline of traffic data visualization, provides an overview of related data processing techniques, and summarizes existing methods for depicting the temporal, spatial, numerical, and categorical properties of traffic data.

mentioning

confidence: 99%

A Survey of Traffic Data Visualization

Chen

Guo

2015

254

137

“…Data mining lies in the intersection of artificial intelligence, machine learning, statistics, and database systems. Representative methods of data mining include clustering, classification, summarization, abnormality detection and regression analysis [17], [18]. While data mining is relevant to statistical analysis, its main goal is to discover unknown and even unexpected models from data and compute the model parameters [19].…”

Section: Analyzing Social Transportation Datamentioning

confidence: 99%

Big Data for Social Transportation

Zheng

Chen

et al. 2016

215

Big data for social transportation brings us unprecedented opportunities for resolving transportation problems for which traditional approaches are not competent and for building the next-generation intelligent transportation systems. Although social data have been applied for transportation analysis, there are still many challenges. First, social data evolve with time and contain abundant information, posing a crucial need for data collection and cleaning. Meanwhile, each type of data has specific advantages and limitations for social transportation, and one data type alone is not capable of describing the overall state of a transportation system. Systematic data fusing approaches or frameworks for combining social signal data with different features, structures, resolutions, and precision are needed. Second, data processing and mining techniques, such as natural language processing and analysis of streaming data, require further revolutions in effective utilization of real-time traffic information. Third, social data are connected to cyber and physical spaces. To address practical problems in social transportation, a suite of schemes are demanded for realizing big data in social transportation systems, such as crowdsourcing, visual analysis, and task-based services. In this paper, we overview data sources, analytical approaches, and application systems for social transportation, and we also suggest a few future research directions for this new social transportation field.

“…The method then leverages this information and uses it to predict the most likely location at any given time in the future. Zhou et al [23] proposed a "semi-lazy" approach that builds models on the fly by using dynamically selected reference trajectories. The advantages of this proposed model are that the target trajectories to be predicted are known before the models are built, and thus, it can derive accurate prediction models with an acceptable delay based on a small number of selected reference trajectories.…”

Section: Related Workmentioning

confidence: 99%

A Self-Adaptive Parameter Selection Trajectory Prediction Approach via Hidden Markov Models

Qiao

Shen

et al. 2015

219

Trajectory prediction of objects in moving objects databases (MODs) has garnered wide support in a variety of applications and is gradually becoming an active research area. The existing trajectory prediction algorithms focus on discovering frequent moving patterns or simulating the mobility of objects via mathematical models. While these models are useful in certain applications, they fall short in describing the position and behavior of moving objects in a network-constraint environment. Aiming to solve this problem, a hidden Markov model (HMM)-based trajectory prediction algorithm is proposed, called Hidden Markov model-based Trajectory Prediction (HMTP). By analyzing the disadvantages of HMTP, a self-adaptive parameter selection algorithm called HMTP * is proposed, which captures the parameters necessary for real-world scenarios in terms of objects with dynamically changing speed. In addition, a density-based trajectory partition algorithm is introduced, which helps improve the efficiency of prediction. In order to evaluate the effectiveness and efficiency of the proposed algorithms, extensive experiments were conducted, and the experimental results demonstrate that the effect of critical parameters on the prediction accuracy in the proposed paradigm, with regard to HMTP * , can greatly improve the accuracy when compared with HMTP, when subjected to randomly changing speeds. Moreover, it has higher positioning precision than HMTP due to its capability of self-adjustment.