Autonomous Wearable Personal Training System with Real-Time Biofeedback and Gesture User Interface

In this paper we present an autonomous wearable personal training system, which includes a gesture user interface and real-time biofeedback. The system consists of a wearable processing device and one or more body-attached inertial sensors with 3-axis accelerometer and 3-axis gyroscope. In our experiments we employ inertial sensors of a smartphone attached to the body. The system processes sensor data and provides users with real-time audio feedback. The operation of the system is driven by user gestures. Gestures are defined user´s body movements that are detected through their characteristic inertial sensor responses. As an example of such a system, we have designed the application for golf swing error detection and correction. The application helps golfers correct the unwanted head movements during the golf swing. The gesture-driven user interface is able to detect the beginning and the end of the swing, swing phases, and the data recording interval. The application guides users to assume the correct setup position. During the golf swing, it provides them with the real-time audio feedback, signalling head movement errors. After the swing, the application indicates head movement errors and draws 3D head movement diagrams. Field test results show that our system is an efficient tool for the detection and correction of head movement errors during the golf swing.