Feasibility of Onboard Smartphones for Railway Track Geometry Estimation: Sensing Capabilities and Characterization

Abstract

The performance and sensitivity of smartphone sensors developed rapidly in recent years. Due to their accessibility and low costs compared to other industrial solutions, the use of smartphone sensors has become more and more common. In this article, the validity and reliability of a smartphone application in railway track geometry estimation are tested on a conventional rail line. This work focused on Galaxy S-series smartphones of Samsung and proposes an evaluation of the onboard sensing capabilities of their inertial sensors with the comparison of synchronous measurements by a multi-functional Track Recording Vehicle equipped with a contactless Track Geometry- (TGMS), and a Vehicle Dynamic Measuring System (VDMS). The raw accelerometer recordings showed a high-degree correlation with VDMS in both signal magnitude and waveform. The accuracy of gyroscope angular tracking in heading and pitching angle calculation was in the range of 0.2°–0.6°, which allowed the acceptable estimation of the central angle and the radius of horizontal curves. Based on the kinematic analysis techniques, the roll flexibility coefficient of the vehicle was determined, which allowed calculating the cross-level and the twist of the track. Furthermore, the local extreme values of the roll-rate gyro correlate with the isolated track geometry defects of the track twist gathered by TRV’s TGMS. Despite its limitations, the application of smartphones represents a prospective technological opportunity to explore new approaches and support rail asset management.