A computer simulation model of a springboard and a diver was developed to
investigate diving takeoff techniques in the forward and the reverse groups. The
springboard model incorporated vertical, horizontal and rotational movements based on
experimental data. The diver was modelled as an eight-segment link system with torque
generators acting at the metatarsal-phalangeal, ankle, knee, hip and shoulder joints.
Wobbling masses were included within the trunk, thigh and shank segments to allow for
soft tissue movement. The foot-springboard interface was represented by spring-dampers
acting at the heel, ball and toes of the foot. The model was personalised to an elite diver so
that simulation output could be compared with the diver's own performance. Kinematic
data of diving performances from a one-metre springboard were obtained using high speed
video and personalised inertia parameters were determined from anthropometric
measurements. Joint torque was calculated using a torque / angle / angular velocity
relationship based on the maximum voluntary torque measured using an isovelocity
dynamometer. Visco-elastic parameters were determined using a subject-specific angledriven
model which matched the simulation to the performance in an optimisation process.
Four dives with minimum and maximum angular momentum in the two dive groups were
chosen to obtain a common set of parameters for use in the torque-driven model. In the
evaluation of the torque-driven model, there was good agreement between the simulation
and performance for all four dives with a mean difference of 6.3%. The model was applied
to optimise for maximum dive height for each of the four dives and to optimise for
maximum rotational potential in each of the two dive groups. Optimisation results suggest
that changing techniques can increase the dive height by up to 2.0 cm. It was also
predicted that the diver could generate rotation almost sufficient to perform a forward three
and one-half somersault tuck and a reverse two and one-half somersault tuck.