Abstract
The horizontal curve is one of the focal points of roadway safety because this curve plays a critical role in transitioning vehicles between tangent roadway sections; moreover, car crashes are frequently concentrated on horizontal curves despite their disproportionate length in the road network. As a critical safety property of horizontal curves, superelevation is crucial to vehicle safety because it counteracts the lateral acceleration produced in vehicles when they travel the curves. Despite the emergence of several sensing-based methods in recent years, labor-intensive and time-consuming manual superelevation evaluation is often carried out by transportation agencies because the newer methods usually demand expensive equipment and complicated operations. Transportation agencies are in urgent need of low-cost, reliable alternatives to improve their data collection practices. This paper proposes an automated superelevation measurement method using inexpensive mobile devices. The proposed method integrates and processes sensing data from a mobile device and derives superelevation by using fundamental vehicle kinematics at a horizontal curve. Kalman filtering–based noise reduction, regression-based radius computation, and complementary-filtering-based rolling angle computation methods are introduced to achieve accurate results despite low-frequency, noisy signals from the inexpensive devices. An experimental test on SR-2 in Georgia demonstrates that the proposed method delivers results with accuracies comparable to those of a lidar-based method. A case study of high friction surface treatment site selection using a ball bank indicator shows that the proposed method is a promising alternative for transportation agencies to achieve low-cost yet reliable data collection for safety analysis and improvement.
Get full access to this article
View all access options for this article.