Map Data Generation Using GPS Density for Missing Road Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for inferring road networks from GPS traces are prone to errors due to atmospheric and satellite interference, leading to false positive detections of road segments, especially when there are few GPS traces, which can result in incorrect navigation and ETA calculations.
Innovation Solution
Utilize Smoothed Particle Hydrodynamics (SPH) analysis to generate cells from GPS points, determine road networks based on GPS point density, and use a deep neural network for semantic segmentation to classify pixels as roads or non-roads, followed by comparing the generated network with a reference map to identify missing roads.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If GPS traces are used directly to infer road networks, then road network detection is possible, but GPS errors lead to false positive detections
Solution Approach 1:
The patent introduces an intermediary processing system that uses multiple data sources (GPS traces, map matching, road network data) and sophisticated algorithms to mediate between raw GPS data and final road network inference, filtering out false positives while preserving true road detections
Solution Approach 2:
The system dynamically adjusts detection parameters and thresholds based on data quality, density, and confidence levels. By changing parameters adaptively rather than using fixed thresholds, the system maintains high detection capability while reducing false positives in different geographic contexts
2Measurement precision
If spatial-linear clustering is used to infer road segments, then road segments can be detected from GPS traces, but Type I and Type II errors increase when GPS traces are sparse
Solution Approach 1:
The patent combines multiple detection methods and data sources (GPS clustering, map matching, existing road network data, point-of-interest data) into an integrated system. This merging approach allows the system to compensate for sparse GPS traces by leveraging complementary information from other sources, reducing both Type I and Type II errors
3Difficulty of detecting and measuring
If conventional intersection detection is used, then intersections can be identified, but GPS errors create false positive signals leading to false positive roads
Solution Approach 1:
The system implements feedback mechanisms where detected intersections and road segments are continuously validated against multiple data sources and confidence thresholds. False positives are identified and corrected through iterative validation processes that compare detections with map matching results, existing road networks, and spatial consistency checks
Data Source
AI summary
Aspects concern a method for generating map data, the method including: generating of a plurality of cells of a real space area, each cell representing a two-dimensional sub-space of the real space area, wherein each of the cells comprise at least one global positioning system (GPS) point, and wherein the real space area comprises a road network; determining a number of GPS points for each cell of the plurality of cells; determining a road network based on the numbers of GPS points using a Smoothed Particle Hydrodynamics (SPH) analysis, wherein the cells of the plurality of cells are particles of the SPH analysis and the number of GPS points per cell of the plurality of cells are a scalar of each cell of the plurality of cells in the SPH analysis; and comparing the road network determined using the SPH analysis with the road network of the real space area.