Related Experiment Video
Updated: Jul 8, 2025

Author Spotlight: Exploring Breathing Techniques and Digital Solutions for Enhancing Running Performance
Published on: September 27, 2024
By cyclists, for cyclists: Road grade and elevation estimation from crowd-sourced fitness application data
Elmira Berjisian1, Alexander Bigazzi1, Hamed Barkh1
1Department of Civil Engineering, University of British Columbia, Vancouver, British Columbia, Canada.
Crowd-sourced GPS data can accurately estimate road grade for walking and cycling analysis. This method is more reliable than elevation models and offers a scalable solution for active transportation planning.
Area of Science:
- Geospatial analysis
- Transportation engineering
- Active transportation research
Background:
- Road grade significantly impacts walking and cycling behavior and outcomes.
- Precise road grade data is scarce, hindering travel information and analysis.
- Crowd-sourced GPS data offers a potential solution for generating road grade datasets.
Purpose of the Study:
- To examine the accuracy of crowd-sourced GPS data for estimating roadway grade profiles.
- To validate and modify existing elevation estimation methods for road grade analysis.
- To develop network-wide road grade datasets for enhanced travel analysis.
Main Methods:
- External validation of an elevation estimation method using field surveying data.
- Modification and evaluation of methods for road grade estimation.
- Aggregation of crowd-sourced GPS data using partition-around-medoid clustering, Savitzky-Golay filter, and DBSCAN.
- Implementation with an average of 150 GPS traces per location.
Main Results:
- Crowd-sourced GPS data provides relatively accurate road grade estimates.
- Grade estimates are more reliable than elevation estimates compared to alternative data sources.
- The most accurate method achieved a root mean square error (RMSE) of 1% road grade.
Conclusions:
- A modest amount of crowd-sourced GPS data can generate accurate road grade estimates.
- This method is superior to low-resolution elevation models but less accurate than LiDAR-derived data.
- Findings enable the incorporation of precise road grade information into street networks for active transportation analysis.
More Related Videos
Related Concept Videos
Profile Leveling and Cross Sections
Elevation of Intermediate Points on Vertical Curves
Design Example: Maintaining Level of an Embankment
Design Example: Identifying the Locations of Monuments in the Field Using Global Positioning System Device
Differential Leveling
Methods of Obtaining Topography

