A New Method for Identifying Hazardous Road Locations Using GPS and Accelerometer

Kristian Hegner Reinau, Camilla Sloth Andersen, Niels Agerholm

Research output: Contribution to book/anthology/report/conference proceedingArticle in proceedingResearchpeer-review


Identification of hazardous road locations is of key importance to traffic safety and has traditionally relied on accident statistics. Due to a declining quality of statistics on accidents in many countries, a need exist for new methods to identify hazardous road locations. This article presents a new method relying on GPS data and accelerometer data. The method developed in this article is superior to previous methods since it utilizes acceleration measurements to identify both a deceleration and a significant drop in speed by integrating acceleration measurements over time in a moving time-window. This calculation is combined with measurements of jerks to distinguish between planned decelerations and unplanned decelerations, which occurs when drivers react to hazardous situations. Finally, using GPS data, the unplanned decelerations identified are linked to specific locations. The method is tested and validated through a case study of Aalborg city, where HRLs identified on the basis of accident statistics are compared to the HCLs identified with the new method on a dataset of more than 36.531 hours of driving tracked with accelerometer and GPS, containing 131 million GPS and 1.1 billion acceleration measurements.
Original languageEnglish
Title of host publicationProceedings of 23rd World Congress on Intelligent Transport Systems : Melbourne 2016
Number of pages12
Publication date2016
Publication statusPublished - 2016
Event23rd ITS World Congress: ITS - Enhancing Liveable Cities and Communities - Melbourne Convention and Exhibition Centre, Melbourne, Australia
Duration: 10 Oct 201614 Oct 2016


Conference23rd ITS World Congress
LocationMelbourne Convention and Exhibition Centre


Dive into the research topics of 'A New Method for Identifying Hazardous Road Locations Using GPS and Accelerometer'. Together they form a unique fingerprint.

Cite this