Road Geometry Generation from Sparse Mobile Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining road geometry, especially for converging and diverging roadways, face challenges in accurately using mobile device data due to difficulties in assigning data points to specific branches, leading to potential false curvatures and logistical issues in updating global road systems.
Innovation Solution
A method that analyzes heading and trajectory information from mobile device data points to identify and group data points associated with particular road branches, using trajectory lines and angle determination to accurately assign data points to converging or diverging road sections, thereby eliminating false curvatures and improving road geometry determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual mapping methods are used to determine road geometry, then accuracy of road geometry can be maintained, but significant resources and time are required for mapping and updating road systems
Solution Approach 1:
The system uses mobile devices carried by users during normal travel to automatically collect position and trajectory data. The road geometry mapping performs itself through self-service by utilizing the movement data of everyday travelers, eliminating the need for dedicated mapping expeditions while maintaining accuracy through aggregated real-world travel paths
Solution Approach 2:
Instead of physically sending mapping personnel to collect data, the system creates digital copies of travel trajectories from mobile device positions. These trajectory copies are then processed to extract road geometry information, replacing physical mapping expeditions with digital data replication and analysis
2Productivity
If mobile device data points are used to determine road geometry, then resource requirements are reduced, but difficulty in assigning data points to specific converging or diverging branches increases
Solution Approach 1:
The system segments the road network into distinct branches at convergence and divergence points. By dividing the complex road geometry into separable branches and analyzing trajectory patterns specific to each branch, the system can accurately assign data points to their respective branches even in complex interchange scenarios
Solution Approach 2:
The system adds temporal dimension to the spatial trajectory data by analyzing the sequence and timing of position data points. This dimensional enhancement allows differentiation between vehicles taking different branches at intersections by examining when and how trajectories diverge or converge, resolving the ambiguity of assigning spatial data points to specific branches
3Area of stationary object
If traditional road mapping methods are used, then comprehensive coverage of global road systems can be achieved, but logistical difficulties increase due to the magnitude and number of roadways
Solution Approach 1:
The system uses mobile devices that serve multiple functions - their primary function for personal use while simultaneously serving as data collection devices for road mapping. This multi-functionality eliminates the need for specialized mapping equipment and personnel, reducing logistical complexity while achieving comprehensive global road system coverage through the universal presence of mobile devices
Solution Approach 2:
The system creates digital replicas of road geometry from mobile device trajectory copies, replacing the need for physical presence and manual mapping activities across global road systems. This digital copying approach eliminates complex logistics of sending personnel worldwide while maintaining comprehensive coverage through automated data collection from everyday travel
Data Source
Figure 1
Figure 2
Figure 3A
AI summary
Road geometries may be determined from sparse data by receiving mobile device data points comprising data indicating positions of vehicles having traveled on a roadway of a geographic area, the roadway involving a convergence or divergence of road branches for the roadway. At least one trajectory angle for a particular mobile device data point of the mobile device data points may be determined using at least one trajectory line connecting the particular mobile device data point to an adjacent mobile device data point, and mobile device data points may be grouped based on the trajectory angles.