The ai supervisor for the effective treadmill training system of rehabilitation and exercise

Abstract

This paper proposes the AI Supervisor which controls the treadmill speed effectively like an expert such as personal trainer or physical therapist based on real-time sensor data and physical information on the user, and AI decision making. It makes a decision to control the speed of a treadmill during exercise or rehabilitation by measuring the heart rate. The decision is processed by the Deep Neural Network (DNN) with a dataset of 8 people, the accurate decision rate is 94.6%.