T-drive: driving directions based on taxi trajectories

By Jing Yuan, Yu Zheng, Chengyang Zhang, Wenlei Xie, Xing Xie, Guangzhong Sun, Yan Huang. ACM SIGSPATIAL International Workshop on Advances in Geographic Information Systems, 2010

1,285 citations recorded by the source corpus.

GPS-equipped taxis can be regarded as mobile sensors probing traffic flows on road surfaces, and taxi drivers are usually experienced in finding the fastest (quickest) route to a destination based on their knowledge. In this paper, we mine smart driving directions from the historical GPS trajectories of a large number of taxis, and provide a user with the practically fastest route to a given destination at a given departure time. In our approach, we propose a time-dependent landmark graph, where a node (landmark) is a road segment frequently traversed by taxis, to model the intelligence of taxi drivers and the properties of dynamic road networks. Then, a Variance-Entropy-Based Clustering approach is devised to estimate the distribution of travel time between two landmarks in different time slots. Based on this graph, we design a two-stage routing algorithm to compute the practically fastest route. We build our system based on a real-world trajectory dataset generated by over 33,000 taxis in a period of 3 months, and evaluate the system by conducting both synthetic experiments and in-the-field evaluations. As a result, 60-70% of the routes suggested by our method are faster than the competing methods, and 20% of the routes share the same results. On average, 50% of our routes are at least 20% faster than the competing approaches.

Subfields: Ocean Engineering, Transportation, Signal Processing

View Paper: https://www.semanticscholar.org/paper/f3528f9c7a6d5ff3f6cc4a01bff6271efefd3d34

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Ocean Engineering

In Transportation

In Signal Processing

Citations in the corpus

A Survey on Trajectory Data Management, Analytics, and Learning (2020, background): “Since trajectory datasets such as T-drive [192] can easily contain millions of points, the R-tree must manage an enormous number of maximum bounding rectangles (MBR), which have prohibitive memory costs in practice.”
Constructing popular routes from uncertain trajectories (2012, background): “The work [14] mainly inferred fastest routes from historical trajectories.”
Trajectory Data Mining (2015, methodology): “After the map-matching process, T-Drive regards the top-k road segments frequently traversed by taxicabs as landmark nodes (i.e., the red points shown in Figure 19(a)).”
Prediction of urban human mobility using large-scale taxi traces and its applications (2012, background): “[13] provided navigation services by extracting the experience of taxi drivers from historical taxi GPS footprints.”
Time-Dependent Trajectory Regression on Road Networks via Multi-Task Learning (2013, background): “(Yuan et al. 2010) built a time-dependent landmark graph by mining from historical traffic data for fast route computation.”
Citywide traffic congestion estimation with social media (2015, methodology): “Previous researches can be roughly categorized into tra ﬃ c modeling on individual roads [8, 11, 14] and on a road network [13, 18, 23].”
Traffic prediction in a bike-sharing system (2015, background): “Yuan et al. [23][24] presented a Cloud-based system that computes the quickest driving routes based on traffic conditions and driver behaviors.”
Graph Neural Network for Traffic Forecasting: A Survey (2021, background): “T-drive [241]: This dataset contains a large number of”
Using smart card data to extract passenger's spatio-temporal density and train's trajectory of MRT system (2012, methodology): “Regarding to estimating real time traﬃc information on urban road network, the methodology for estimating the travel time and spatio-temporal traﬃc density on road surfaces was introduced by using GPS trajectory data [13, 14].”
iBAT: detecting anomalous taxi trajectories from GPS traces (2011, background): “[26] designed PROCAB and T-Drive, respectively, providing driving direction guidance by leveraging taxis’ GPS traces.”
Driving with knowledge from the physical world (2011, methodology): “Also, a rough route indicating the key directions reduces the search area on a road network and enables parallel computing when performing the detailed routing [26].”
T-Drive: Enhancing Driving Directions with Taxi Drivers' Intelligence (2013, methodology): “Index Terms—Spatial databases and GIS, data mining, GPS trajectory, driving directions, driving behavior Ç”
Computing Urban Traffic Congestions by Incorporating Sparse GPS Probe Data and Social Media Data (2017, methodology): “…congestion estimation on an arterial transportation network can be roughly categorized into trafﬁc modeling on individual roads [Helbing 2001; Mu˜noz et al. 2003; Porikli and Li 2004; de Fabritiis et al. 2008] and on a road network [Pattara-Atikom et al. 2006; Yuan et al. 2010; Shang et al. 2014].”
Urban Traffic Prediction from Spatio-Temporal Data Using Deep Meta Learning (2019, methodology): “We obtain taxi flows from TDrive dataset [25], which contains a large amount of taxicab trajectories from Feb.”
Travel time estimation of a path using sparse trajectories (2014, background): “Based on the trajectories generated by a large number of taxis, [20] builds a landmark graph, where a node (entitled a landmark) is a road segment frequently traveled by taxis and an edge denotes the aggregation of taxis’ commutes between two landmarks.”