Compression of GPS Trajectories using Autoencoders

TL;DR

This paper introduces an LSTM autoencoder approach for compressing GPS trajectories, demonstrating superior performance over traditional algorithms like Douglas-Peucker in terms of accuracy and flexibility across different datasets.

Contribution

The paper presents a novel LSTM autoencoder method for GPS trajectory compression, outperforming traditional algorithms in accuracy and allowing lossy reconstruction for enhanced benefits.

Findings

Outperforms Douglas-Peucker in discrete Fréchet distance and dynamic time warping.

Effective on both gaming and real-world GPS datasets.

Offers advantages through lossy point reconstruction.

Abstract

The ubiquitous availability of mobile devices capable of location tracking led to a significant rise in the collection of GPS data. Several compression methods have been developed in order to reduce the amount of storage needed while keeping the important information. In this paper, we present an lstm-autoencoder based approach in order to compress and reconstruct GPS trajectories, which is evaluated on both a gaming and real-world dataset. We consider various compression ratios and trajectory lengths. The performance is compared to other trajectory compression algorithms, i.e., Douglas-Peucker. Overall, the results indicate that our approach outperforms Douglas-Peucker significantly in terms of the discrete Fr\'echet distance and dynamic time warping. Furthermore, by reconstructing every point lossy, the proposed methodology offers multiple advantages over traditional methods.