A survey of sensor fusion algorithms for sport and health monitoring applications

Abstract

The development of Micro-Electro-Mechanical Systems based inertial sensors allows the implementation of a large number of commercial and scientific solutions for, health and sports monitoring, navigation systems and aeronautical/aerospace engineering applications. Inertial sensors widespread the so called embedded systems also prompted for the development of algorithms to cope with their inherent inaccuracy. These sensors, with small form factor and low cost, present an advantage for computing three-dimensional space object orientation. Also radio frequency communication devices are frequently associated with inertial sensor and embedded systems. Radio frequency communication protocols allow immediate data transfer and processing without cumbersome cabling. This work presents a survey of a number of relevant sensor fusion algorithms. The identified algorithms are implemented and its performance tested. Also a Yaskawa Motoman robotic arm is used to implement a set of predefined movements that granted the mandatory test accuracy and repeatability conditions. A Bluetooth Low Energy device has been implemented to transfer the inertial sensor data to a PC computer running a Windows operating system. The inertial sensors have been implemented with a MPU-9150 Inertial Measurement Unit (IMU) that includes an accelerometer, a gyroscope and a compass.