Prediction of Ground Reaction Forces from Pressure Measurements

By Stephan Odenwald

Challenges occur, if GRF should be evaluated without floor-integrated force measurement platforms.

Measuring kinetic parameters such as ground reaction forces (GRF) is a basic requirement in gait analyses. The scope of this project was to measure kinetic data under laboratory and field conditions by using pressure measurement insoles. The main focus of the project was the acquisition of vertical GRF during walking and running. Vertical GRF was predicted by means of artificial neuronal networks (ANN) that used plantar pressure distribution measurements from the insoles as input values. In total, eleven handball and basketball players took part in the validation study. To develop a suitable ANN for the prediction of vertical GRF the networks were initially trained with GRF data measured by a force plate and adjusted with simultaneously measured pressure data. Following to the training, the most suitable ANN was selected and validated with further GRF and pressure data. Since the validation was successfully, it can be stated that the applied ANN had the ability to predict vertical GRF during walking and running by means of plantar pressure measurements. Furthermore the comparison between laboratory and field measurements resulted also in high congruities between the basic kinetic parameters. To sum up, artificial neuronal networks in combination with pressure insoles are a promising method for the acquisition of ground reaction forces in real-life measurements.

Leave a comment


  • No comments found