On the Arm Swing Model During Human Walking for Wrist-Worn Rotational Electret Energy Harvester

In this report, a machine-learning model for a wrist-worn rotational electret energy harvester (REEH) is presented. By using random forest regression (RFR) on a motion capture dataset, an arm swing model is developed to predict the output power during human walking. Compared to conventional linear regression, the present model shows better agreement with the experimental data of seven subjects wearing a REEH with a wireless output power measurement system.