Taxonomy of Proactive Detection Methods of Drunk Driving for Enhancing Traffic Safety

Drunk driving continues to be a substantial public health issue, leading to a multitude of accidents and deaths on a global scale. Conventional techniques for identifying drunk driving, such as breathalyzers and field sobriety tests, tend to be reactive rather than proactive. This paper surveys the existing methods for detecting drunk driving to enhance traffic safety. The technologies studied can proactively detect impaired driving behaviors by combining machine learning (ML) approaches with real-time data processing. The study outlines various classifications based on measured metrics and driving behavior.