Context‐aware Android applications through transportation mode detection techniques

Abstract

In this paper, we study the problem of how to detect the current transportation mode of the user from the smartphone sensors data, because this issue is considered crucial for the deployment of a multitude of mobility‐aware systems, ranging from trace collectors to health monitoring and urban sensing systems. Although some feasibility studies have been performed in the literature, most of the proposed systems rely on the utilization of the GPS and on computational expensive algorithms that do not take into account the limited resources of mobile phones. On the opposite, this paper focuses on the design and implementation of a feasible and efficient detection system that takes into account both the issues of accuracy of classification and of energy consumption. To this purpose, we propose the utilization of embedded sensor data (accelerometer/gyroscope) with a novel meta‐classifier based on a cascading technique, and we show that our combined approach can provide similar performance than a GPS‐based classifier, but introducing also the possibility to control the computational load based on requested confidence. We describe the implementation of the proposed system into an Android framework that can be leveraged by third‐part mobile applications to access context‐aware information in a transparent way. Copyright © 2016 John Wiley & Sons, Ltd.

Custom Dual Transportation Mode Detection By Smartphone Devices Exploiting Sensor Diversity

“…The set of classes considered is composed of WALKING, CAR, STILL, TRAIN and BUS. This follows common practices in literature [3] [15].…”

“…However, it suffers from heavy battery consumption and scarce accuracy in indoor environments or urban canyons [19], due to fading and multipath signals which lower the GPS accuracy. Moreover, it is unable to correctly classify transportation modes with similar speeds [3]. Sensor-based approaches are often based on Machine Learning (ML) techniques and on training set of classified instances.…”

“…For location-awareness, the magnetometer showed good performance in locating the user through fingerprinting maps [11], while the barometer has been used for locating the altitude from the ground level [4]. For TMD, the most used sensor is certainly the accelerometer, which provides the best trade-off between the accuracy of the activity recognized and the energy consumption [3] [13] [9]. External-source based approaches enhance the robustness of TMD algorithms by extra-source information that are relative to the current user location.…”

“…We also highlight that there is a wide literature on the algorithms for TMD, including both the application/evaluation of well-known ML techniques (e.g. Random Forest is largely used in [3][15] [18]), and the design of novel recognition schemes [6]. However, at the best of our knowledge, no prior study addressed the problem of Dual TDM (i.e.…”

“…on how to infer the transportation mode experienced by the user, on the basis of the smartphone sensors. Many popular mobile AP-Plications (APP) include functionalities of automatic Transportation Mode Detection (TMD) [3], for instance related to the execution of automated actions in response of detected locations and mobility-aware events; we cite, among other, the popular If This Than That (IFTTT) framework [1]. In large-scale urban scenarios, TMD techniques are also employed to enable the seamless gathering of mobility traces on a voluntary basis, avoiding the need of external sensing infrastructures, and at the same time minimizing the annoyance for the users [19].…”