FOR THE DEGREE OF LICENTIATE OF ENGINEERING in Machine and Vehicle Systems Volunteer Muscle Activity in Dynamic Events Input Data for Human Body Models

Human body models (HBMs) are virtual human surrogates used to predict kinematic and injury responses during motor vehicle crashes. In recent years, active musculature has been incorporated into HBMs for enhanced biofidelity in simulated emergency scenarios, in particular low-severity crashes and pre-crash situations, where occupant responses are influenced by muscle tension. Development and validation of HBMs that simulate neuromuscular control requires information on muscle activation patterns and contraction levels for different loading levels and directions. This information can be acquired by measuring muscle activity in volunteers with electromyography in replicated pre-crash events. This thesis investigates occupant responses in various pre-crash type braking scenarios and multidirectional perturbations. Muscle activity was measured in volunteers in the following scenarios; maximum voluntary braking, autonomous braking with standard seatbelt, autonomous braking with reversible pre-tensioner activated 200 ms before braking, and seated perturbation in multiple directions without restraint. Muscle activity and forward displacement during autonomous braking was influenced by type of restraint system and role (passenger vs. driver). Pre-tensioning the seatbelt caused decreased forward displacement as well as increased startle like muscle activity in some volunteers. Active HBMs that model the startle reflex can elucidate its effect on injury risk in the crash phase. The difference in posture between drivers and passengers resulted in decreased upper extremity and increased lower back muscle activity for passengers and more forward displacement. Active HBMs validated against the data presented here can be used to further assess the difference between the two occupant roles and to aid the optimisation of safety systems for each group. The spatial tuning patterns generated from multidirectional perturbation showed variable activation amplitudes and preferred directions for the neck muscles. Implementing muscle and direction specific activation schemes in active HBMs might result in better prediction of the head and neck responses. The research outcomes provide data sets for active HBM validation in pre-crash braking events and the development and validation of omnidirectional models. Further studies that identify occupant muscle responses are needed. Measuring muscle activity during a pre-crash steering manoeuvre or during a realistic visual threat to identify the muscle responses following a startle reflex would support the advancement of future omnidirectional models and startle reflex control methods.