Distance-Based Bucket Processing for HD Map Slope and Curvature Updates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current systems face challenges in efficiently processing and aggregating large volumes of high-frequency, high-quality sensor data from autonomous vehicles into meaningful high-definition map datasets, particularly in determining reliable slope and curvature values for road mapping.
Innovation Solution
A method and apparatus that map-match vehicle data sessions with map links, divide them into distance-based buckets, and process these buckets for variance statistics to determine reliability information for acceleration slope and curvature values, enabling updates to maps based on this information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If voluminous sensor datasets are processed directly without segmentation, then complete map coverage is achieved, but processing efficiency and reliability determination become intractable
Solution Approach 1:
The patent divides map links into multiple distance-based buckets (e.g., 10-meter intervals) to segment the continuous road network into discrete, manageable units. This segmentation enables parallel processing of individual buckets while maintaining complete road coverage, resolving the contradiction between processing efficiency and reliability accuracy by making the data tractable without losing information integrity
2Device complexity
If map links are divided into multiple distance-based buckets, then processing complexity is reduced, but data integration and consistency become more challenging
Solution Approach 1:
The patent implements feedback mechanisms where variance statistics from processed buckets are used to determine reliability information, which then feeds back into the map updating process. This feedback loop ensures that only reliable slope and curvature data (those with low variance across multiple sensor readings) are integrated into the HD map, maintaining data consistency across all bucket segments
3Reliability
If variance statistics processing is applied to all buckets, then reliability information is improved, but computational resources are consumed
Solution Approach 1:
The patent applies variance statistics processing locally to each individual bucket rather than globally across the entire map. This local quality approach allows computational resources to be distributed and utilized efficiently in parallel across multiple buckets, improving reliability determination while avoiding the prohibitive computational cost of global processing
Data Source
AI summary
An approach is provided for converting voluminous sensor datasets into manageable distance based buckets for slope and curvature map updates. The approach involves causing, at least in part, a map-matching of one or more data sessions with one or more map links, wherein the one or more data sessions represent one or more location data packages associated with a vehicle. The approach also involves causing, at least in part, a division of the one or more map links into one or more buckets at one or more intervals. The approach further involves processing and/or facilitating a processing of the one or more buckets for variance statistics to determine reliability information for acceleration slope values, acceleration curvature values, or a combination thereof calculated from the one or more location data packages. The approach also involves causing, at least in part, an updating of one or more maps based, at least in part, on the reliability information.