Page 19

EDNE MAY 2015

Sensors which allows a direct comparison of end user devices. This article introduces the concept of sensor fusion within a smart phone context. It discusses how sensor fusion software is used to improve overall accuracy and introduces a test method including performance result measurement on a number of flagship smart phones. The fusion library described uses accelerometer, magnetometer and gyroscope sensor signals to compensate for each others’ shortcomings and provides highly accurate, reliable and stable orientation data. Let’s look at the strengths and weaknesses of these critical devices and how they compensate for each others drawbacks. The orientation of an object describes how it is placed in the 3-D space and typically the orientation is given relative to a frame of reference specified by a coordinate system. At least three independent values as part of a 3-dimensional vector are needed to describe true orientation. All points of the body change their position during a rotation except for those lying on the rotation axis. The magnetometer A magnetometer is highly sensitive to interfering local magnetic fields and distortions, which result in errors in the calculated magnetic heading. The gyroscope can be used to detect such interference and heading changes where no rotation is registered. Sensor fusion can then accurately compensate for this by giving more weight to the gyroscope data than to that of the magnetometer. The authors continue by outlining the properties of each of the magnetometer, the accelerometer and the gyroscope, and how they are combined in real-world systems; they conclude by showing how error data can be derived to characterise systems performance. Download PDF of Article Find Sensor fusion on EETsearch 18 EDN Europe | MAY 2015 www.edn-europe.com


EDNE MAY 2015
To see the actual publication please follow the link above