Enrichment of OpenStreetMap Data Completeness with Sidewalk Geometries Using Data Mining Techniques

Area of Science:

• Geographic Information Systems (GIS)
• Data Mining
• Urban Planning

Background:

• Wheelchair routing requires detailed sidewalk geometry data, which is often missing in current mapping databases like OpenStreetMap.
• The CAP4Access project aims to enhance OpenStreetMap for wheelchair accessibility.

Purpose of the Study:

• To present a modified methodology for constructing sidewalk geometries using data mining techniques.
• To enrich OpenStreetMap with detailed sidewalk data for improved wheelchair routing.

Main Methods:

• Utilized multiple GPS traces from wheelchair users during urban travel experiments.
• Applied data mining techniques to construct sidewalk geometries.
• Validated the method with a case study in Heidelberg, Germany, comparing results to an official reference dataset.

Main Results:

• The constructed sidewalk network achieved 96% overlay with the official reference dataset.
• A low Root Mean Square Error (RMSE) of 0.93 m indicates high positional accuracy.
• Demonstrated the feasibility of enriching OpenStreetMap for wheelchair routing.

Conclusions:

• The proposed methodology effectively constructs accurate sidewalk geometries for wheelchair routing.
• Enriching OpenStreetMap with this data significantly improves its utility for accessibility.
• Future research can further refine the method and expand its application.