Determining trip and travel mode from GPS and accelerometer data

Abstract

The use of Global Positioning Systems (GPS) and/or accelerometers to identify trips and transportation modes such as walking, running, bicycling or motorized transportation has been an active goal in multiple disciplines such as Transportation Engineering, Computer Science, Informatics and Public Health. The purpose of this study was to review existing methods that determined trip and travel mode from raw Global Positioning System (GPS) and accelerometer data, and test a select group of these methods. The study had three specific aims: (1) Create a systematic review of existing literature that explored various methods for determining trip and travel mode from GPS and/or accelerometer data, (2) Collect a convenience sample of subjects who were assigned a GPS and accelerometer unit to wear while performing and logging travel bouts consisting of walking, running, bicycling and driving, (3) Replicate selected method designs extracted from the systematic review (aim 1) and use subject data (aim 2) to compare the methods. The results were be used to examine which methods are effective for various modes of travel.