Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Car-Following Models: A Multidisciplinary Review

Tianya Zhang, Ph. D., Peter J. Jin, Ph. D., Sean T. McQuade, Ph. D., Alexandre Bayen, Ph. D., Benedetto Piccoli

TL;DR

This work provides a comprehensive, multidisciplinary synthesis of car-following models, spanning theory-based, learning-based, and knowledge-driven approaches. It introduces a novel taxonomy, contrasts classical models (e.g., IDM, OVM, Gipps, Newell) with data-driven and AI-driven paradigms, and surveys their applications, datasets, and stability properties. The paper highlights the strengths and limitations of each paradigm, advocates hybrid and physics-informed strategies, and outlines future directions such as world-model predictive control, hierarchical planning, multimodal sensing, and SSL for robust, interpretable autonomous driving. Together, these insights guide researchers and practitioners in designing CF models that are accurate, scalable, and safe in mixed-traffic and connected environments.

Abstract

Car-following (CF) algorithms are crucial components of traffic simulations and have been integrated into many production vehicles equipped with Advanced Driving Assistance Systems (ADAS). Insights from the model of car-following behavior help us understand the causes of various macro phenomena that arise from interactions between pairs of vehicles. Car-following models encompass multiple disciplines, including traffic engineering, physics, dynamic system control, cognitive science, machine learning, and reinforcement learning. This paper presents an extensive survey that highlights the differences, complementarities, and overlaps among microscopic traffic flow and control models based on their underlying principles and design logic. It reviews representative algorithms, ranging from theory-based kinematic models, Psycho-Physical Models, and Adaptive cruise control models to data-driven algorithms like Reinforcement Learning (RL) and Imitation Learning (IL). The manuscript discusses the strengths and limitations of these models and explores their applications in different contexts. This review synthesizes existing researches across different domains to fill knowledge gaps and offer guidance for future research by identifying the latest trends in car following models and their applications.

Car-Following Models: A Multidisciplinary Review

TL;DR

This work provides a comprehensive, multidisciplinary synthesis of car-following models, spanning theory-based, learning-based, and knowledge-driven approaches. It introduces a novel taxonomy, contrasts classical models (e.g., IDM, OVM, Gipps, Newell) with data-driven and AI-driven paradigms, and surveys their applications, datasets, and stability properties. The paper highlights the strengths and limitations of each paradigm, advocates hybrid and physics-informed strategies, and outlines future directions such as world-model predictive control, hierarchical planning, multimodal sensing, and SSL for robust, interpretable autonomous driving. Together, these insights guide researchers and practitioners in designing CF models that are accurate, scalable, and safe in mixed-traffic and connected environments.

Abstract

Car-following (CF) algorithms are crucial components of traffic simulations and have been integrated into many production vehicles equipped with Advanced Driving Assistance Systems (ADAS). Insights from the model of car-following behavior help us understand the causes of various macro phenomena that arise from interactions between pairs of vehicles. Car-following models encompass multiple disciplines, including traffic engineering, physics, dynamic system control, cognitive science, machine learning, and reinforcement learning. This paper presents an extensive survey that highlights the differences, complementarities, and overlaps among microscopic traffic flow and control models based on their underlying principles and design logic. It reviews representative algorithms, ranging from theory-based kinematic models, Psycho-Physical Models, and Adaptive cruise control models to data-driven algorithms like Reinforcement Learning (RL) and Imitation Learning (IL). The manuscript discusses the strengths and limitations of these models and explores their applications in different contexts. This review synthesizes existing researches across different domains to fill knowledge gaps and offer guidance for future research by identifying the latest trends in car following models and their applications.
Paper Structure (28 sections, 22 equations, 4 figures, 7 tables)

This paper contains 28 sections, 22 equations, 4 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Theory-based Models
  3. Kinematic Models
  4. First-Order Kinematic Models
  5. Second-Order Kinematic Models
  6. Psycho-physical Models
  7. Adaptive Cruise Control Model
  8. Linear Model
  9. Nonlinear Model
  10. Model Predictive Control (MPC)
  11. Learning-based Models
  12. Supervised Learning Models
  13. Machine Learning Model
  14. Deep Learning Model
  15. Imitation Learning Models
  16. ...and 13 more sections

Figures (4)

  • Figure 1: Car-Following and Microscopic Traffic Flow Properties.
  • Figure 2: Proposed Taxonomy of Car Following Models (CFMs) and Representative Algorithms. Kinematic Models are divided into first-order and second-order subcategories. Knowledge-driven Large GenAI Models are a revolutionary category, providing new insights and solutions with commonsense.
  • Figure 3: The Historical Timeline of Reviewed Master Models and Key Variants.
  • Figure 4: Navigating the Literature with Reviewed Master Models Using OVM and IDM as Examples