Towards Attention-Based Convolutional Long Short-Term Memory for Travel Time Prediction of Bus Journeys.

Jianqing Wu,Chen Cai,Jun Shen,Qiang Wu

doi:10.3390/s20123354

Abstract

Travel time prediction is critical for advanced traveler information systems (ATISs), which provides valuable information for enhancing the efficiency and effectiveness of the urban transportation systems. However, in the area of bus trips, existing studies have focused on directly using the structured data to predict travel time for a single bus trip. For state-of-the-art public transportation information systems, a bus journey generally has multiple bus trips. Additionally, due to the lack of study on data fusion, it is even inadequate for the development of underlying intelligent transportation systems. In this paper, we propose a novel framework for a hybrid data-driven travel time prediction model for bus journeys based on open data. We explore a convolutional long short-term memory (ConvLSTM) model with a self-attention mechanism that accurately predicts the running time of each segment of the trips and the waiting time at each station. The model is more robust to capture long-range dependence in time series data as well.

Highlights

The usage of intelligent transportation systems (ITSs) is motivated in a significant part by passenger increases and sustainable development [1,2]
This paper proposes a hybrid model that applies the ConvLSTM network with an attention mechanism to explore a suitable model for the bus journey time prediction on open data
We applied two standard metrics to evaluate the performance of running time prediction and waiting time prediction, including root mean square errors (RMSEs) and mean absolute errors (MAEs)

Summary

Introduction

The usage of intelligent transportation systems (ITSs) is motivated in a significant part by passenger increases and sustainable development [1,2]. Developing sustainable and intelligent transportation applications operate and manage real-time and historical data efficiently, which has become an increasingly important yet challenging task. It plays a vital role in achieving the main objectives of ITS, which include accessibility and mobility, environmental sustainability and economic development [3,4]. With the advent of artificial intelligence (AI), machine learning and expert system-based paradigms have driven the development of society and the steady growth of the economy. Deep learning can discover patterns in complex data sets, which could not be found via conventional methods.

Objectives

Methods

Results

Discussion

Conclusion