Control-oriented modelling of occupants in frontal impacts

Abstract

This paper presents manageable, mathematical models of a vehicle occupant and a belt restraint system. These low-order models are relevant in the development of controlled restraint systems, which aim at lowering injury criteria by real-time control of the occupant motion. The models can be employed for control design and real-time injury prediction, the main components of controlled restraint systems. Several low-order models are constructed with first principles of physics and through knowledge obtained from a sensitivity analysis of validated, complex occupant models. The biomechanical responses of the low-order models, related to neck and thoracic injury criteria are validated with the results of the complex models. They are found to be valid for 5th, 50th and 95th percentile Hybrid III dummies in a range of frontal impact scenarios. The conclusion of this study is that thoracic and neck injury criteria in frontal impact can be accurately assessed with relatively simple occupant models, which are required for real-time control of injury-related biomechanical responses.