COMPUTER SIMULATION IN SPORTS BIOMECHANICS

Abstract

Experimental studies of sports technique are inherently problematical. In competitive situations biomechanical variables are constrained to lie near optimal values and this results in a small range of values for each variable. As a consequence there is often no statistically significant relationship between predictor and performance variables. In the training environment intervention is possible and the ranges of variable values can be increased. It is difficult, however, to change just one technique variable without affecting others and there is the possibility that technique becomes so perturbed that it is no longer representative of an unconstrained performance. Theoretical studies employing linked segment models of the human body do not suffer from these experimental difficulties. With such a model it is a simple matter to determine the influence of just one variable on performance. The difficulty is in developing a sufficiently detailed model that will represent the key features of the sports movement. The complexity of a model depends on the application. If the aim is to gain basic insights into the mechanics of the movement then a rather simple model may be sufficient. If the aim is to determine optimum technique then a more complex subject-specific model will be required. A drawback of oversimplified models is that they cannot match actual performances with great accuracy. As a consequence there is the possibility that a key element will have been omitted from the model and the insights gained into the sports movement may be erroneous. For complex models of the human body there are the problems of obtaining accurate representations of joint constraints, segmental inertias, muscle performance, and elastic structures within the body. Evaluating a model using actual performance data is a non-trivial necessity if there is to be confidence in the model predictions. The applications of simulation models to sports movements provides a means of understanding the underlying mechanics, analysing competitive performances, determining optimum techniques, and providing a simulated performance environment to assist in the learning of complex control. Examples of simulation models of increasing complexity will be used to illustrate the many applications of simulation with examples drawn from jumping activities in athletics and gymnastics.