scientific papers

Near Real-Time Transportation Mode Detection Based on Accelerometer Readings

November 8, 2016

by Jožef Stefan Institute, Ljubljana, Slovenia

This paper describes a method for automatic transportation mode detection based on smartphone sensors. Our approach is designed to work in real time, as it only requires 5s of sensor readings for detection. Because we are using an accelerometer rather than a GPS signal, the method uses less battery power and is therefore more user and phone friendly. For mode detection, we use multiple support vector machine models that enable us to distinguish between multiple modes (bus, train, car). Before the classication, raw measurements are pre-processed in order to cancel out the constant acceleration that is caused by the force of gravity. The results of the paper are promising and are based on collected training data from approximately 20 hours of driving on trains and public buses in Ljubljana.