Discovering Car-following Dynamics from Trajectory Data through Deep Learning

TL;DR

This paper introduces a deep learning framework that discovers interpretable mathematical models of car-following behavior directly from trajectory data, combining symbolic regression with variable selection to produce parsimonious expressions.

Contribution

The study presents a novel deep symbolic regression approach with variable intersection selection for uncovering interpretable car-following models from trajectory data.

Findings

Successfully learned several car-following dynamics models

Enhanced interpretability through parsimonious mathematical expressions

Demonstrated the effectiveness of penalties in guiding model discovery

Abstract

This study aims to discover the governing mathematical expressions of car-following dynamics from trajectory data directly using deep learning techniques. We propose an expression exploration framework based on deep symbolic regression (DSR) integrated with a variable intersection selection (VIS) method to find variable combinations that encourage interpretable and parsimonious mathematical expressions. In the exploration learning process, two penalty terms are added to improve the reward function: (i) a complexity penalty to regulate the complexity of the explored expressions to be parsimonious, and (ii) a variable interaction penalty to encourage the expression exploration to focus on variable combinations that can best describe the data. We show the performance of the proposed method to learn several car-following dynamics models and discuss its limitations and future research directions.