Map Data Generation Using Point Cloud Height Filtering
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current map development and navigation services face challenges in accurately determining the location and geometry of road objects like tunnels, bridges, and roundabouts due to insufficient GPS signal strength, leading to incomplete and non-real-time navigation data.
Innovation Solution
A system and method that utilize point cloud data to generate georeferenced top-down raster images, filter height data, and apply semantic image segmentation or computer vision techniques to identify boundaries and geometry of road objects, integrating GPS probe data for speed limit determination and dynamic navigation assistance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If manual geo-coding is used to update map data with road objects, then map features can be added, but the process is not scalable and does not reflect real-time status
Solution Approach 1:
The system enables automated map updating by having vehicles self-collect GPS probe data and point cloud data during normal operation, which is then processed to automatically update map features. This eliminates manual geo-coding and enables both real-time accuracy and scalability.
Solution Approach 2:
The patent replaces manual mechanical geo-coding processes with automated computational systems that process GPS probe data and point cloud data using algorithms and machine learning models to automatically identify and update road objects.
2Measurement precision
If GPS signals are used to determine vehicle location in tunnels and bridges, then navigation data can be obtained, but GPS signal strength is insufficient leading to inaccurate location determination
Solution Approach 1:
The system introduces point cloud data as an intermediary to supplement GPS probe data. By combining GPS coordinates with point cloud geometry data and processing them together through algorithms, the system can accurately determine vehicle location and road object geometry even when GPS signals are weak or unavailable.
Solution Approach 2:
The patent creates a composite data structure by integrating GPS probe data with point cloud data, combining the strengths of both data sources to achieve accurate location and geometry determination in environments where either data source alone would be insufficient.
3Measurement precision
If point cloud data is processed without height filtering, then all data points are available, but processing efficiency decreases and accuracy is reduced
Solution Approach 1:
The system extracts and separates relevant data points from the complete point cloud by applying height filtering. This removes irrelevant ground-level points and keeps only the elevated points that represent road objects like bridges and overpasses, simplifying subsequent boundary identification processing.
Solution Approach 2:
The patent applies different processing strategies to different portions of the point cloud data based on their height characteristics. By filtering based on height thresholds, the system treats elevated road objects differently from ground-level features, improving boundary identification accuracy for overhead structures.
Data Source
AI summary
A method, a system, and a computer program product may be provided for generating map data associated with one or more objects in a region. The method includes receiving point cloud data associated with the region. The method further includes generating a georeferenced top down raster image of the region, based on the point cloud data. The georeferenced top down raster image is indicative of a top surface of each of the one or more objects. The method further includes determining boundary data of the top surface of each of the one or more objects, based on the georeferenced top down raster image and generating the map data associated with the one or more objects, based on the boundary data of the top surface of each of the one or more objects.