Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: Apr 15, 2026

Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research

Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research

Published on: December 18, 2020

Prevention of Motorcycle-Car Door Collisions by Using a Deep-Learning-Based Automatic Braking Assistance System.

Yaojung Shiao^1,2, Tan-Linh Huynh¹

¹Department of Vehicle Engineering, National Taipei University of Technology, Taipei 10608, Taiwan.

Sensors (Basel, Switzerland)

|April 14, 2026

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Three-Dimensional Human Posture Recognition by Extremity Angle Estimation with Minimal IMU Sensor.

Sensors (Basel, Switzerland)·2024

Same author

Accuracy Improvement of Automatic Smoky Diesel Vehicle Detection Using YOLO Model, Matching, and Refinement.

Sensors (Basel, Switzerland)·2024

Same author

Exercise Condition Sensing in Smart Leg Extension Machine.

Sensors (Basel, Switzerland)·2022

See all related articles

This study introduces a low-cost, camera-based automatic braking system for motorcycles using deep learning to detect opening car doors. The system enhances motorcycle safety by optimizing braking performance and alerting riders to prevent common, fatal collisions.

Area of Science:

Road Safety Engineering
Artificial Intelligence in Transportation
Motorcycle Safety Systems

Background:

Motorcycle-car door collisions are frequent and often fatal.
High costs of radar sensors limit current motorcycle safety advancements.
A need exists for affordable, effective collision avoidance systems for motorcycles.

Purpose of the Study:

To develop an inexpensive, camera-based automatic braking assistance system for motorcycles.
To enhance motorcycle braking performance and collision avoidance using deep learning.
To detect car door states and optimize braking response.

Main Methods:

Implemented a deep learning model (YOLOv12s) for car door state detection.
Designed safety mechanisms to adjust braking intensity and front braking ratio.

Keywords:

YOLOv12s object detection model automatic braking assistance system braking intensity car door states front braking ratio initial braking speed preventable accidents

More Related Videos

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

Related Experiment Videos

Last Updated: Apr 15, 2026

Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research

Tactile Vibrating Toolkit and Driving Simulation Platform for Driving-Related Research

Published on: December 18, 2020

Design and Analysis for Fall Detection System Simplification

Design and Analysis for Fall Detection System Simplification

Published on: April 6, 2020

Integrated time-to-collision, rider braking action, and initial speed into the system's logic.

Main Results:

YOLOv12s achieved high accuracy in car door detection (90.5% precision, 80.6% recall, 87.8% mAP).
The system demonstrated optimal braking performance by adjusting intensity (0-100%) and front braking ratio.
At 60 km/h, a 45% front braking ratio reduced braking distance significantly compared to 20% and 60% ratios.

Conclusions:

The proposed camera-based system offers a feasible, low-cost solution for motorcycle collision avoidance.
The system responds within a critical 0.5-second time window.
Further research is needed to enhance model robustness with diverse data and implement real-time physical actuators.