Improved Crash Detection Algorithm for Vehicle Crash Detection
Road accidents are one of the major causes of injuries, loss of life, and economic damage worldwide, with many accidents occurring due to careless or distracted driving. Therefore, developing an accurate and reliable vehicle crash detection system is essential for improving road safety and providing timely emergency response. Although several crash detection algorithms have been proposed, many of them are designed for specific situations and may not perform effectively under different road conditions or collision types. This paper presents an improved vehicle crash detection algorithm that uses sensor data from the x, y, and z axes along with time information to accurately identify crash events in real-world driving conditions. The proposed algorithm analyze the vehicle's motion and impact characteristics to improve detection accuracy while reducing false crash alerts. It also distinguishes between driving mode and parking mode, ensuring that unnecessary crash detections caused by parking movements or minor disturbances are avoided. The algorithm was tested under various driving and parking scenarios, and the results showed that it can reliably detect crash events while minimizing false detections. In addition, the proposed method helps optimize the storage capacity of the vehicle's black box system by recording and storing only important crash event videos. This approach improves the efficiency, reliability, and practical use of vehicle crash detection systems and can contribute to enhanced road safety and faster emergency assistance.
Authors
- Sharvari Shinde
- Shreya Kadam
- Sandhya Deshpande
- Yash Ratwadkar
Institutions
- K J Somaiya Medical College (IN)
Publication Details
- Journal
- International Journal of Latest Technology in Engineering Management & Applied Science
- Published
- 2026-11-21
- DOI
- https://doi.org/10.51583/ijltemas.2026.151100007
- Primary Topic
- IoT and GPS-based Vehicle Safety Systems
- Type
- article
- Field-Weighted Citation Impact
- 0.00