Abstract
An optimization of the fluctuating punted kick is carried out using a genetic algorithm. It was assumed that there are five objective functions and nine control variables. The nine control variables, which are under the control of the kicker, determine the launch conditions. Since the control variables are too many to enable the optimal values to be determined manually, the optimization study is clearly necessary in order to find the optimal punted kick. Carrying out multi-objective optimization, the trade-offs between the objective functions and the control variables could be visualized using Self-Organizing Maps. It was found that the higher the spin rate at launch the greater the number of fluctuations; however, the hang time becomes shorter. Moreover, it was visualized how the control variables affected the objective functions by bubble charts.
