Bayesian Deep Learning Approach for Real-time Lane-based Arrival Curve Reconstruction at Intersection using License Plate Recognition Data

TL;DR

This study proposes a Bayesian deep learning approach for real-time lane-based arrival curve reconstruction, in which the lane choice patterns and uncertainties of link-based arrivals are both characterized and the relationship between partially observed link-based arrivals and lane-based arrivals is captured.

Abstract

The acquisition of real-time and accurate traffic arrival information is of vital importance for proactive traffic control systems, especially in partially connected vehicle environments. License plate recognition (LPR) data that record both vehicle departures and identities are proven to be desirable in reconstructing lane-based arrival curves in previous works. Existing LPR databased methods are predominantly designed for reconstructing historical arrival curves. For real-time reconstruction of multi-lane urban roads, it is pivotal to determine the lane choice of real-time link-based arrivals, which has not been exploited in previous studies. In this study, we propose a Bayesian deep learning approach for real-time lane-based arrival curve reconstruction, in which the lane choice patterns and uncertainties of link-based arrivals are both characterized. Specifically, the learning process is designed to effectively capture the relationship between partially observed link-based arrivals and lane-based arrivals, which can be physically interpreted as lane choice proportion. Moreover, the lane choice uncertainties are characterized using Bayesian parameter inference techniques, minimizing arrival curve reconstruction uncertainties, especially in low LPR data matching rate conditions. Real-world experiment results conducted in multiple matching rate scenarios demonstrate the superiority and necessity of lane choice modeling in reconstructing arrival curves.

Figures (8)

• Figure 1: Problem statement and the relationship between cumulative departure and arrival curves.
• Figure 2: Overview of the proposed lane choice learning-based arrival curve reconstruction framework.
• Figure 3: Left: Classical neural network, each weight has a fixed value. Right: Bayesian neural network, each weight is represented by a distribution.
• Figure 4: Illustration of the reparameterization trick.
• Figure 5: Visualization of the test site layout and signal timing.