Capacitive-Type Pressure Sensor for Classification of the Activities of Daily Living

In order to operate a gait rehabilitation device, it is necessary accurately classify the states appearing in activities of daily living (ADLs). case force sensing resistors (FSRs), which are often used as pressure sensors analysis, desirable replace them with other because their low durability. present study, capacitive-type sensors, an alternative FSRs, were developed, and performance was evaluated. addition, timed up go test performed measure ground reaction healthy individuals, machine learning technique applied calculated biosignal parameters for classification five types ADLs. The evaluation results showed that sensor thermoplastic polyurethane (substrate dielectric layer material) multiwall carbon nanotubes (conductive layer) has sufficient sensitivity durability use analysis sensor. Moreover, when overlapping filter four-layer long short-term memory (LSTM) or five-layer LSTM model developed motion classification, precision greater equal 95%, unstable errors did not occur. Therefore, ADLs algorithm this study applied, expected can be completed within time range does affect control device.