The effect of joint compliance within rigid whole-body computer simulations of impacts

In high impact human activities, much of the impact shock wave is dissipated through internal body structures, preventing excessive accelerations from reaching vital organs. Mechanisms responsible for this attenuation, including lower limb joint compression and spinal compression have been neglected in existing whole-body simulation models. Accelerometer data on one male subject during drop landings and drop jumps from four heights revealed that peak resultant acceleration tended to decrease with increasing height in the body. Power spectra contained two major components, corresponding to the active voluntary movement (2 Hz 14 Hz) and the impact shock wave (16 Hz 26 Hz). Transfer functions demonstrated progressive attenuation from the MTP joint towards the C6 vertebra within the 16 Hz 26 Hz component. This observed attenuation within the spine and lower-limb joint structures was considered within a rigid body, nine-segment planar torque-driven computer simulation model of drop jumping. Joints at the ankle, knee, hip, shoulder, and mid-trunk were modelled as non-linear spring-dampers. Wobbling masses were included at the shank, thigh, and trunk, with subject-specific biarticular torque generators for ankle plantar flexion, and knee and hip flexion and extension. The overall root mean square difference in kinetic and kinematic time-histories between the model and experimental drop jump performance was 3.7%, including ground reaction force root mean square differences of 5.1%. All viscoelastic displacements were within realistic bounds determined experimentally or from the literature. For an equivalent rigid model representative of traditional frictionless pin joint simulation models but with realistic wobbling mass and foot-ground compliance, the overall kinetic and kinematic difference was 11.0%, including ground reaction force root mean square differences of 12.1%. Thus, the incorporation of viscoelastic elements at key joints enables accurate replication of experimentally recorded ground reaction forces within realistic whole-body kinematics and removes the previous need for excessively compliant wobbling masses and/or foot-ground interfaces. This is also necessary in cases where shock wave transmission within the simulation model must be non-instantaneous.