Browsing by Subject "vehicle crash testing"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Analysis of Potential Co-Benefits for Bicyclist Crash Imminent Braking Systems
    (IEEE, 2017-10) Good, David H.; Krutilla, Kerry; Chien, Stanley; Li, Lingxi; Chen, Yaobin; Electrical and Computer Engineering, School of Engineering and Technology
    In the US, the number of traffic fatalities has had a long term downward trend as a result of advances in the crash worthiness of vehicles. However, these improvements in crash worthiness do little to protect other vulnerable road users such as pedestrians or bicyclists. Several manufacturers have developed a new generation of crash avoidance systems that attempt to recognize and mitigate imminent crashes with non-motorists. While the focus of these systems has been on pedestrians where they can make meaningful contributions to improved safety [1], recent designs of these systems have recognized mitigating bicyclist crashes as a potential co-benefit. This paper evaluates the performance of one system that is currently available for consumer purchase. Because the vehicle manufacturer does not claim effectiveness for their system under all crash geometries, we focus our attention on the crash scenario that has the highest social cost in the US: the cyclist and vehicle on parallel paths being struck from behind. Our analysis of co benefits examines the ability to reduce three measures: number of crashes, fatalities, and a comprehensive measure for social cost that incorporates morbidity and mortality. Test track simulations under realistic circumstances with a realistic surrogate bicyclist target are conducted. Empirical models are developed for system performance and potential benefits for injury and fatality reduction. These models identify three key variables in the analysis: vehicle speed, cyclist speed and cyclist age as key determinants of potential co-benefits. We find that the evaluated system offers only limited benefits for any but the oldest bicycle riders for our tested scenario.
  • Thumbnail Image
    Item
    Predicting Hazardous Driving Events Using Multi-Modal Deep Learning Based on Video Motion Profile and Kinematics Data
    (IEEE, 2018-11) Gao, Z.; Liu, Y.; Zheng, J. Y.; Yu, R.; Wang, X.; Sun, P.; Computer and Information Science, School of Science
    As the raising of traffic accidents caused by commercial vehicle drivers, more regulations have been issued for improving their safety status. Driving record instruments are required to be installed on such vehicles in China. The obtained naturalistic driving data offer insight into the causal factors of hazardous events with the requirements to identify where hazardous events happen within large volumes of data. In this study, we develop a model based on a low-definition driving record instrument and the vehicle kinematic data for post-accident analysis by multi-modal deep learning method. With a higher camera position on commercial vehicles than cars that can observe further distance, motion profiles are extracted from driving video to capture the trajectory features of front vehicles at different depths. Then random forest is used to select significant kinematic variables which can reflect the potential crash. Finally, a multi-modal deep convolutional neural network (DCNN) combined both video and kinematic data is developed to identify potential collision risk in each 12-second vehicle trip. The analysis results indicate that the proposed multi-modal deep learning model can identify hazardous events within a large volumes of data at an AUC of 0.81, which outperforms the state-of-the-art random forest model and kinematic threshold method.