Document Type

Thesis - Open Access

Award Date


Degree Name

Master of Science (MS)

Department / School

Civil and Environmental Engineering

First Advisor

Jonathan Wood


Curve Calculation, Curve Identification, EKF, Horizontal curve, Mobile Computing, Smartphone


Smartphones and other portable personal devices that integrate global positioning systems, Bluetooth Low Energy, and advanced computing technologies have become more accessible due to affordable prices, product innovation, and people’s desire to be connected. As more people own these devices, there are greater opportunities for data acquisition in Intelligent Transportation Systems, and for vehicle-to-infrastructure communication. Horizontal curves are a common factor in the number of observed roadway crashes. Identifying locations and geometric characteristics of the horizontal curves plays a critical role in crash prediction and prevention, and timely curve warnings save lives. However, most states in the US face a challenge to maintain detailed and highquality roadway inventory databases for low volume rural roads due to the laborintensive and time-consuming nature of collecting and maintaining the data. This thesis proposes two smartphone applications C-Finder and C-Alert, to collect two-lane road horizontal curves data (including radius, superelevation, length, etc.), collect this data for transportation agencies (providing a low-cost alternative to mobile asset data collection vehicles), and for warning drivers of sharp horizontal curves, respectively. C-Finder is capable of accurately detecting horizontal curves by exploiting an unsupervised K-means machine learning technique. Butterworth low pass filtering was applied to reduce sensor noise. Extended Kalman filtering was adopted to improve GPS accuracy. Chord method-based radius computation, and superelevation estimation were introduced to achieve accurate and robust results despite of the low-frequency GPS and noisy sensor signals obtained from the smartphone. C-Alert applies BLE technology and a head-up display (HUD) to track driver speed and compare vehicle position with curve locations in a real-time fashion. Messages can be wirelessly communicated from the smartphone to a receiving unit through BLE technology, and then displayed by HUD on the vehicle’s front windshield. The field test demonstrated that C-Finder achieves high curve identification accuracy, reasonable accuracy for calculating curve radius and superelevation compared to the previous road survey studies, and C-Alert indicates relatively high accuracy for speeding warning when approaching sharp curves.

Library of Congress Subject Headings

Rural roads -- Safety measures.
Curves in engineering.
Traffic safety.
Mobile apps.


Includes bibliographical references (pages 44-50)



Number of Pages



South Dakota State University



Rights Statement

In Copyright