Evaluation of the offline classification error of human locomotion modes using virtual force-sensing resistor data

Active lower limb assistive devices are intended to contribute to human movement activities such as locomotion. While some devices are capable to function among a variety of terrains, wide instrumentation sets have been usually applied to minimize classification errors for locomotion mode recognition, even if this potentially bias user comfort. The objective of this study was to find a combination of parameters for the offline classification of five steady-state locomotion modes based on kinetic data from three discrete areas on a pressure insole to reduce classification error. Parametric tests were used to compare the effect of different combinations on signal processing schemes on the classification error of linear discriminant analysis classifiers. Results showed that the best overall classifier had an average error of 7.85±4.76%. This method presents an insight about the use of a reduced number of sensors for gait classification on the development of comfortable lower limb assistive devices.