Car-Following Model with Automatic Reaction Delay Estimation: An Attention-Based Ensemble Learning Methodology

Abstract

Car-following behavior is a vital traffic phenomenon in the process of vehicle driving. For modeling the car-following behavior, it is crucial to capture the reaction delay for balancing with safety and comfort, but it is generally ignored in existing works. This work proposes a car-following model based on attention-based ensemble learning to automatically capture the reaction delay from driving data and better depict the traffic flow characteristics. The model integrates a data-driven model and a theory-driven model, and a weight computation method is proposed to combine the advantage of these two different models. In detail, an encoder-decoder model and attention mechanism are employed to capture the reaction delay from driving data. Extensive experiments show that the proposed model could balance safety with comfort and help avoid unsafe driving behavior.

An Attention-Based Car-Following Model Based on Fused Data

“…To further validate the performance of the proposed ABF-CF model, we compare the safety indicator TH m and accuracy indicator MSE a of the ABF-CF model with several existing models, including the IDM, the GRU based car-following model, the Seq2Seq-LSTM (S-LSTM) [5], the AEL-CF [9], and the Gipps-BPNN [8]. The results are shown in Table 6.…”

“…Since theory-driven and data-driven car-following models have their own advantages and disadvantages, some scholars have attempted to develop model-data fusion car-following models expecting to achieve an outstanding performance. One of the most straightforward approaches is to combine theory-driven and data-driven car-following models via linear combination forecasting methods [8], [9]. In particular, Li et al proposed a fusion car-following model with the adaptive Kalman filter algorithm to integrate a long-short term memory neural network with the Intelligent Driver Model (IDM) model [10].…”