MobileSense - Sensing Modes of Transportation in Studies of the Built Environment

Abstract

We discuss a study we conducted using the Mobile Sensing Platform and GPS information to record the activities and locations of 53 subjects, each of whom collected data for one week. Using this data we are developing methods for combining activity inference from sensors to infer the mode of transportation, label significant locations, and extract trips. Although our data collection was conducted using a non-consumer sensor platform, we discuss how our methods can translate onto existing mobile platforms such as the iPhone and Nokia N95, and how these platforms will enable us to study larger populations to draw more concrete conclusions about the relationship between the urban environment and people’s activities.