Verification of the Peak Time Approach for Detection of Step Initiation using the UTRCEXO

Abstract

We were able to detect the step initiation for the Unmanned Technology Research Center Exoskeleton before visible movements occurred during the peak time approach. Detection of the step initiation is important for the rapid onset of assistance with the exoskeleton operator's movement. Many previous studies have attempted to detect the step initiation more rapidly using the precedence walking assistance mechanism with electromyography, or the shadow walking assistance mechanism with the heel-off or toe-off time. In this paper, we detect the step initiation and implement the precedence walking assistance mechanism using the peak time approach. In particular, we detect the vertical ground reaction forces before visible movements occur, which is more reliable, simpler and faster than the previous approaches. We also present insole-type force sensing resistors based on the peak time approach that are used in force plates that can be applied to the Unmanned Technology Research Center Exoskeleton to detect similar events, such as the ground reaction force events, and the step initiation. With the insole-type force sensing resistors, the Unmanned Technology Research Center Exoskeleton can not only detect step initiation before visible movements occur, but can also implement the precedence walking assistance mechanism for step initiation without using any bio-signals.