Hill Climbing Algorithm for HD Map Construction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for creating high-definition (HD) maps, such as those using aerial or satellite imaging, are expensive and require human labeling, making them inefficient and costly for widespread use, especially for autonomous vehicle navigation.
Innovation Solution
A method utilizing a hill climbing algorithm to construct HD maps by receiving multi-layer probability density bitmaps from sensors of multiple vehicles, recursively conducting a hill climbing search to create lines representing lane lines, and generating an HD map without human intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If aerial or satellite imaging is used to create HD maps, then map accuracy is improved, but cost and complexity increase significantly
Solution Approach 1:
The patent creates a virtual copy of the road map by aggregating and processing sensor data from multiple vehicles to generate a probability density bitmap, which represents the road geometry without requiring physical aerial imaging or satellite photography
Solution Approach 2:
The patent replaces the mechanical/aerial imaging system with a computational approach using sensor data fusion and probability density mapping, substituting physical image capture with algorithmic map construction from vehicle sensor inputs
2Measurement precision
If aerial or satellite imaging is used to create HD maps, then map accuracy is improved, but time and labor requirements increase due to human labeling
Solution Approach 1:
The system performs self-service by automatically processing sensor data from multiple vehicles through the probability density bitmap method and hill climbing algorithm, eliminating the need for human labeling and manual map construction
Solution Approach 2:
The patent performs preliminary data aggregation and probability density calculation during normal vehicle operation, so that when HD map construction is needed, the processing is already substantially complete, reducing final construction time
3Ease of manufacture
If crowd-sourcing approaches are used to create HD maps, then cost is reduced, but data processing complexity increases
Solution Approach 1:
The patent merges sensor data from multiple vehicles into a single probability density bitmap, combining multiple data sources into one unified representation that simplifies subsequent processing while maintaining accuracy through data fusion
Solution Approach 2:
The patent transforms raw sensor data into a probability density representation, changing the parameter space from individual sensor readings to aggregated probability values, which simplifies the hill climbing algorithm's search process
Data Source
AI summary
A method of creating a high-definition (HD) map of a roadway includes receiving a multi-layer probability density bitmap. The multi-layer probability density bitmap represents a plurality of lane lines of the roadway sensed by a plurality of sensors of a plurality of vehicles. The multi-layer probability density bitmap includes a plurality of points. The method further includes recursively conducting a hill climbing search using the multi-layer probability density bitmap to create a plurality of lines. In addition, the method includes creating the HD map of the roadway using the plurality of lines determined by the hill climbing search.