Open Access

ARTICLE

Time Synchronized Velocity Error for Trajectory Compression

Haibao Jiang¹, Dezhi Han^1,*, Han Liu¹, Jiuzhang Han¹ and Wenjing Nie²

¹College of Information Engineering, Shanghai Maritime University, Shanghai, 201306, China
²College of Foreign Language, Shanghai Maritime University, Shanghai, 201306, China

* Corresponding Author:Dezhi Han. Email:

(This article belongs to the Special Issue: New Trends in Statistical Computing and Data Science)

Computer Modeling in Engineering & Sciences 2022, 130(2), 1193-1219. https://doi.org/10.32604/cmes.2022.017663

Received 28 May 2021; Accepted 20 August 2021; Issue published 13 December 2021

Abstract

Nowadays, distance is usually used to evaluate the error of trajectory compression. These methods can effectively indicate the level of geometric similarity between the compressed and the raw trajectory, but it ignores the velocity error in the compression. To fill the gap of these methods, assuming the velocity changes linearly, a mathematical model called SVE (Time Synchronized Velocity Error) for evaluating compression error is designed, which can evaluate the velocity error effectively, conveniently and accurately. Based on this model, an innovative algorithm called SW-MSVE (Minimum Time Synchronized Velocity Error Based on Sliding Window) is proposed, which can minimize the velocity error in trajectory compression under the premise of local optimization. Two elaborate experiments are designed to demonstrate the advancements of the SVE and the SW-MSVE respectively. In the first experiment, we use the PED, the SED and the SVE to evaluate the error under four compression algorithms, one of which is the SW-MSVE algorithm. The results show that the SVE is less influenced by noise with stronger performance and more applicability. In the second experiment, by marking the raw trajectory, we compare the SW-MSVE algorithm with three others algorithms at information retention. The results show that the SW-MSVE algorithm can take into account both velocity and geometric structure constraints and retains more information of the raw trajectory at the same compression ratio.

---

Keywords

---