Detecting Taxi Speeding from Sparse and Low-Sampled Trajectory Data
Taxis are a major means of public transportation in large cities, and speeding is a common problem among motor vehicles, including taxis. Unless caught by sensors or patrol officers, many speeding incidents go unnoticed, which pose potential threat to road safety. In this paper, we propose to detect speeding behaviors of individual taxis from taxi trajectory data. Such detection results are useful for driver risk analysis and road safety management. However, the taxi trajectory data are geographically sparse and the sample rate is low. Furthermore, existing methods mainly deal with the estimation of collective road speeds whereas we focus on the speeds of individual vehicles. As such, we propose to use a two-fold collective matrix factorization (CMF)-based model to estimate the individual vehicle speed. We have evaluated our method on real-world datasets, and the results show the effectiveness of our method in detecting taxi speeding behaviors.
KeywordsSpeeding Collective matrix factorization Trajectory
This work is supported in part by the Guangdong Pre-national project 2014GKXM054 and the Guangdong Province Key Laboratory of Popular High Performance Computers 2017B030314073.
- 5.Wang, Y., Zheng, Y., Xue, Y.: Travel time estimation of a path using sparse trajectories. In: Proceedings of the 20th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 25–34. ACM (2014)Google Scholar
- 6.Wang, Z., Li, M., Wang, L., Liu, X.: Estimation trajectory of the low-frequency floating car considering the traffic control. Math. Prob. Eng. 2013, 11 (2013)Google Scholar
- 7.Xin, X., Lu, C., Wang, Y., Huang, H.: Forecasting collector road speeds under high percentage of missing data. In: AAAI, pp. 1917–1923 (2015)Google Scholar