Trip Detection with Smartphone-Assisted Collection of Travel Data

This paper puts forward a method that automatically detects trips and trip segments with data on the instantaneous movement attributes of individuals that can be collected automatically by smartphone sensors. The goal is to enhance the accuracy of the data collected through the better identification of single-mode trips and trip segments while minimizing the participant’s involvement and preserving battery life. The proposed method works independently of data from external sources and can be implemented in smartphone applications to enhance the accuracy of the data that are collected and minimize the amount of data that need to be transferred. The method consists of a combination of real-time processing and postprocessing of the data and incorporates a series of rules to clean, split, and merge trips and trip segments, if required. The performance of the model was evaluated in a real-world experiment, in which it achieved an overall accuracy of 97% for the detection of trips from records of daily tracks. The analysis of the results shows that the implementation of the trip detection model increased the proportion of nonmotorized trips detected by 6%. In addition, the implementation of the model increased the accuracy of the data on the duration and the length of the recorded trips.