Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Empirical Performance Evaluation of Lane Keeping Assist on Modern Production Vehicles

Yuhang Wang, Abdulaziz Alhuraish, Shuyi Wang, Hao Zhou

TL;DR

This work provides the first large-scale, real-world evaluation of Lane Keeping Assist (LKA) performance using the OpenLKA dataset, revealing that failures fall into perception, planning, and control categories and that drift increases with road curvature. It quantifies how faded or low-contrast lane markings and sharp curves dominate LKA failures and introduces a dynamic road-geometry–speed model along with a data-driven predictor to assess infrastructure readiness for LKA deployment. The study demonstrates a near-linear relationship between curvature and lane deviation, highlights limitations of current LKA on curved and rural roads, and offers concrete design and policy guidance to align road infrastructure with automated systems. Collectively, these findings support safer autonomous driving through improved infrastructure standards, adaptive LKA control, and targeted rural road interventions.

Abstract

Leveraging a newly released open dataset of Lane Keeping Assist (LKA) systems from production vehicles, this paper presents the first comprehensive empirical analysis of real-world LKA performance. Our study yields three key findings: (i) LKA failures can be systematically categorized into perception, planning, and control errors. We present representative examples of each failure mode through in-depth analysis of LKA-related CAN signals, enabling both justification of the failure mechanisms and diagnosis of when and where each module begins to degrade; (ii) LKA systems tend to follow a fixed lane-centering strategy, often resulting in outward drift that increases linearly with road curvature, whereas human drivers proactively steer slightly inward on similar curved segments; (iii) We provide the first statistical summary and distribution analysis of environmental and road conditions under LKA failures, identifying with statistical significance that faded lane markings, low pavement laneline contrast, and sharp curvature are the most dominant individual factors, along with critical combinations that substantially increase failure likelihood. Building on these insights, we propose a theoretical model that integrates road geometry, speed limits, and LKA steering capability to inform infrastructure design. Additionally, we develop a machine learning-based model to assess roadway readiness for LKA deployment, offering practical tools for safer infrastructure planning, especially in rural areas. This work highlights key limitations of current LKA systems and supports the advancement of safer and more reliable autonomous driving technologies.

Empirical Performance Evaluation of Lane Keeping Assist on Modern Production Vehicles

TL;DR

This work provides the first large-scale, real-world evaluation of Lane Keeping Assist (LKA) performance using the OpenLKA dataset, revealing that failures fall into perception, planning, and control categories and that drift increases with road curvature. It quantifies how faded or low-contrast lane markings and sharp curves dominate LKA failures and introduces a dynamic road-geometry–speed model along with a data-driven predictor to assess infrastructure readiness for LKA deployment. The study demonstrates a near-linear relationship between curvature and lane deviation, highlights limitations of current LKA on curved and rural roads, and offers concrete design and policy guidance to align road infrastructure with automated systems. Collectively, these findings support safer autonomous driving through improved infrastructure standards, adaptive LKA control, and targeted rural road interventions.

Abstract

Leveraging a newly released open dataset of Lane Keeping Assist (LKA) systems from production vehicles, this paper presents the first comprehensive empirical analysis of real-world LKA performance. Our study yields three key findings: (i) LKA failures can be systematically categorized into perception, planning, and control errors. We present representative examples of each failure mode through in-depth analysis of LKA-related CAN signals, enabling both justification of the failure mechanisms and diagnosis of when and where each module begins to degrade; (ii) LKA systems tend to follow a fixed lane-centering strategy, often resulting in outward drift that increases linearly with road curvature, whereas human drivers proactively steer slightly inward on similar curved segments; (iii) We provide the first statistical summary and distribution analysis of environmental and road conditions under LKA failures, identifying with statistical significance that faded lane markings, low pavement laneline contrast, and sharp curvature are the most dominant individual factors, along with critical combinations that substantially increase failure likelihood. Building on these insights, we propose a theoretical model that integrates road geometry, speed limits, and LKA steering capability to inform infrastructure design. Additionally, we develop a machine learning-based model to assess roadway readiness for LKA deployment, offering practical tools for safer infrastructure planning, especially in rural areas. This work highlights key limitations of current LKA systems and supports the advancement of safer and more reliable autonomous driving technologies.
Paper Structure (15 sections, 6 equations, 15 figures)

This paper contains 15 sections, 6 equations, 15 figures.

Table of Contents

  1. Introduction
  2. Background and Dataset
  3. Related Works
  4. The OpenLKA Dataset
  5. Curation of LKA Failure Dataset
  6. Performance Metrics and Evaluation Methods
  7. Empirical LKA Performance Analysis
  8. Perception
  9. Planning
  10. Control
  11. Multiple Factors
  12. Discussions
  13. Implications for Road Geometry Design and Speed Limits
  14. Distilling Empirical Observations into Predictive Insights: Lane Markings and Rural Road Safety
  15. Conclusion

Figures (15)

  • Figure 1: Lane Position Evaluation Metrics
  • Figure 2: Lane Deviation: Curvature is less than 0.001, and the Openpilot's lane detection probability is greater than 0.9. The road curvature is low, the lane detection effect is acceptable, and the significance value reflects the performance under normal conditions to a certain extent.
  • Figure 3: Steering stability under scenarios whose curvature is less than 0.0005 and the lane detection results are greater than 0.95.
  • Figure 4: Low pavement-laneline contrast (white pavement white laneline) causes lane detection failure
  • Figure 5: Common Cases in Laneline Detection Failure
  • ...and 10 more figures