A Sparse Taxi Trajectory Data Recovery and Calibrate Algorithm

Data sparseness is a common problem in many taxi trajectories. After analyzing the characteristics of taxi operation, this paper proposes a sparse taxi trajectory recovery and calibrate algorithm which based on the reference systems (RS) and used heterogeneous data sources. The algorithm increases the number of points in the original trajectory through searching and selecting RS points. Then the trajectory is interpolated and calibrated to improve accuracy and solve sparsity. We test our algorithm by real taxi trajectory data and display visual results in experiment. The all-day trajectory is restored and showed with its high accuracy, feasibility and credibility. It can effectively reduce the candidate paths and recover the taxi trajectory.