Portabelt Sensorsystem för Mätning av Fingrarnas Rörelser - Wearable Sensor based System for Measuring Finger Motions

Abstract

Abstract The hand is humanity’s most used tool and an injury to this body part would be detrimental to both a persons mental and physical well being. If an injury has occurred an important step on the road to recovery is rehabilitation. With the advancements made in technology it is now possible to track the movements of the hand and fingers using different systems. The objective of this thesis was to investigate whether Hall effect sensors could be a suitable choice for such a system. This was done by creating a prototype. The final glove is made out of 14 magnets and 3D Hall effect sensors for measuring the angles of the finger’s joints. These send their data to an Arduino micro controller that packages the data and sends it to a neural network. The purpose of the network is to map the sensor data to joint angles collected from a stereo camera hand tracking system to then be able to predict angles using only the sensor data. The predicted angles are then visualized with a hand model in Unity. The project shows promising results for the use of Hall effect sensors in a motion tracking device, as they produce unique data for different joint angles. However there is much room for improvement, especially when it comes to the camera setup and hand tracking software but also for the optimization of the neural network. This thesis concludes that Hall effect sensors in combination with a neural network and high quality camera data is a viable choice for a finger motion tracking device.