An SVM-Based Method Combining AEB and Airbag Systems to Reduce Injury of Unbelted Occupants

Paper #:
  • 2018-01-1171

Published:
  • 2018-04-03
Abstract:
Autonomous emergency braking (AEB) system can detect the emergency conditions using sensors (e.g., radar and camera) to activate the braking actuator automatically without driver input. However, during the hard braking phase, the crash conditions for restraint system can be easily changed (e.g., vehicle velocity and occupant position), causing out-of-position (OOP) phenomenon, especially for unbelted occupants entering the airbag deployment range, which may lead to more severe injuries than normal position. Therefore, how to design the AEB system and consider the airbag effect simultaneously should be a critical step to reduce injury of unbelted occupants. But few studies have paid attention to the compatibility between AEB and airbag systems for unbelted occupants. This study aims to provide a method that combines AEB and airbag systems to explore the potential injury reduction capabilities for unbelted occupants. By dividing the distance between head and steering wheel into five regions, the possible head position area was obtained for each combination of braking acceleration and time by computational investigation. Taking braking acceleration and time as input features as well as real-time head position area in one of five regions as output label, the support vector machine (SVM) classification model was trained and validated by the obtained data set. The ride-down efficiency of different region was compared, the optimization for maximum delta V reduction was conducted based on SVM model under the prerequisite of guaranteeing high ride-down efficiency and appropriate forward displacement for the occupant. By utilizing the method mentioned above to design the integrated safety systems including AEB and airbag systems, the safety benefits of this approach were finally demonstrated by comparing with the original design.
Access
Now
SAE MOBILUS Subscriber? You may already have access.
Buy
Attention: This item is not yet published. Pre-Order to be notified, via email, when it becomes available.
Select
Price
List
Download
$22.00
Mail
$22.00
Members save up to 36% off list price.
Share
HTML for Linking to Page
Page URL

Related Items

Training / Education
2007-03-01
Event
2018-04-10
Article
2017-07-26
Technical Paper / Journal Article
2011-05-17