SPTTE: A Spatiotemporal Probabilistic Framework for Travel Time Estimation

TL;DR

SPTTE introduces a novel spatiotemporal probabilistic framework utilizing RNN-based Gaussian processes to accurately estimate evolving joint travel time distributions under sparse and uneven data conditions.

Contribution

The paper presents a new probabilistic modeling approach that captures temporal variability of multi-trip travel times with sparse, uneven data using RNN-based Gaussian processes and heterogeneity smoothing.

Findings

SPTTE outperforms existing methods by over 10.13% in accuracy.

The model effectively captures temporal dependencies in travel time data.

Component ablations confirm the importance of each model element.

Abstract

Accurate travel time estimation is essential for navigation and itinerary planning. While existing research employs probabilistic modeling to assess travel time uncertainty and account for correlations between multiple trips, modeling the temporal variability of multi-trip travel time distributions remains a significant challenge. Capturing the evolution of joint distributions requires large, well-organized datasets; however, real-world trip data are often temporally sparse and spatially unevenly distributed. To address this issue, we propose SPTTE, a spatiotemporal probabilistic framework that models the evolving joint distribution of multi-trip travel times by formulating the estimation task as a spatiotemporal stochastic process regression problem with fragmented observations. SPTTE incorporates an RNN-based temporal Gaussian process parameterization to regularize sparse observations and capture temporal dependencies. Additionally, it employs a prior-based heterogeneity smoothing strategy to correct unreliable learning caused by unevenly distributed trips, effectively modeling temporal variability under sparse and uneven data distributions. Evaluations on real-world datasets demonstrate that SPTTE outperforms state-of-the-art deterministic and probabilistic methods by over 10.13%. Ablation studies and visualizations further confirm the effectiveness of the model components.