Authors: Lanxin Zhang, Yongqi Dong, Haneen Farah, Arkady Zgonnikov, Bart van Arem
Published on: December 07, 2023
Impact Score: 8.0
Arxiv code: Arxiv:2312.04610
Summary
- What is new: The study introduces a semi-supervised machine learning method using Hierarchical Extreme Learning Machines (HELM) and Surrogate Safety Measures (SSMs) as input for detecting abnormal driving behaviors.
- Why this is important: The difficulty of detecting abnormal driving behaviors with limited labeled data and the dependence on basic vehicle motion features.
- What the research proposes: A Hierarchical Extreme Learning Machines based semi-supervised ML method that uses partly labeled data and incorporates Surrogate Safety Measures (SSMs) to improve detection.
- Results: The proposed method achieved significant accuracy (99.58%) and F-1 measure (0.9913), outperforming other semi-supervised or unsupervised methods.
Technical Details
Technological frameworks used: Hierarchical Extreme Learning Machines (HELM)
Models used: Semi-supervised ML models
Data used: Large-scale real-world driving data
Potential Impact
Automotive safety technology providers, autonomous vehicle developers, and insurance companies could benefit or be disrupted by these insights.
Want to implement this idea in a business?
We have generated a startup concept here: SafeDrive Analytics.
Leave a Reply