ABSTRACT General Motors (GM), OnStar and the University of Michigan International Center for Automotive Medicine (ICAM) have formed a partnership to investigate and analyze real world rollover crashes involving GM vehicles equipped with rollover sensing technology and rollover-capable roof rail airbag systems. Candidates for the study are initially identified by OnStar, who receive notification of a rollover crash through the vehicle's Automatic Crash Response system. If the customer agrees to participate in the study, medical, vehicle and crash scene information are quickly gathered. This information is then reviewed by the medical and GM engineering communities to provide field relevant learning on injury mechanisms and vehicle system performance in rollover events. This paper provides a detailed review of the field case studies collected to date. INTRODUCTION General Motors began voluntarily implementing rollover sensing technology and rollover-capable roof rail airbag systems in 2005 model year midsize sport utility vehicles. The focus of this technology is to reduce the risk of occupant ejection during a rollover event. According to FARS data from 1992 - 2004, approximately half of the fatalities that occurred in rollover crashes involved complete ejection of the occupant. This does not include fatalities associated with partial ejections [ 1]. Prior to implementing these technologies, laboratory tests were developed to simulate real world rollover conditions. These included a suite of vehicle tests to develop rolloversensor calibrations. These tests were designed to simulate different types of rollover initiations [ 2]. In addition, rollover- capable roof rail airbag systems were developed and assessed for their ability to mitigate the risk of ejection through a series of linear impactor subsystem tests [ 3]. The rigorous laboratory tests developed to assess and implement these new technologies provide confidence that these systems will perform well in the field and ultimately achieve the goal of reducing the risk of ejection for an occupant involved in a real world rollover crash. A key element in confirming that the laboratory approach does in fact translate to real world rollovers is a study of the field data itself. Unfortunately, traditional sources for this real world field information take a significant amount of time to accumulate data and vary with their level of detail in order to assess the performance of a new technology such as rollover sensing and rollover-capable roof rail airbag systems. With the desire for quicker and more in-depth information regarding the performance of GM's rollover systems, an advanced field study was constructed. The objective of this study was to leverage OnStar's ability to identify in real time an event involving an airbag deployment commanded by the rollover sensor, and to partner with the University of Michigan ICAM to obtain critical medical information associated with occupants involved in rollover field events. This advanced field study will assist the medical field to better predict and understand rollover injuries, to better treat them in a timely fashion, and to improve emergency medical response techniques. It will also permit engineers to collect real world rollover data, analyze the information and make Advanced Field Study of Rollover Sensor Equipped Vehicles2011-01-1113 Published 04/12/2011 Stephen J. Cassatta, Anthony G. Melocchi, Huizhen Lu, Lisa A. Stacey and Theron V. Weinberg General Motors Company Stewart C. Wang and Carla J. Kohoyda-Inglis University of Michigan ICAM Copyright © 2011 SAE International doi:10.4271/2011-01-1113Downloaded from SAE International by American Univ of Beirut, Monday, July 30, 2018assessment of the occupant protection system performance. This paper provides the background for this ongoing field study, and details the results from the case studies that have been collected to date. STUDY DESCRIPTION In structuring this field s