Semantic Map Sharding for Lower-Load HD Map Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles require large, high-definition maps that are typically stored as monolithic data files, which are resource-intensive and difficult to upload and process, especially due to their size, leading to high network capacity and processing requirements.
Innovation Solution
Semantic map sharding splits high-definition map data into logical chunks called shards, allowing for efficient storage and processing by generating shard data based on semantic objects and geographical sections, reducing data size and processing load.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If monolithic map data files are used to store high-definition maps for autonomous vehicles, then complete geographical coverage and semantic object information are provided, but network capacity requirements and processing load increase significantly
Solution Approach 1:
The patent divides monolithic map data into multiple shard files organized in a hierarchical directory structure. Each shard contains a subset of semantic objects for a specific geographical section, allowing selective loading and processing. This segmentation reduces the processing load on autonomous vehicles while maintaining access to complete map data when needed.
Solution Approach 2:
The patent extracts and separates semantic objects from complete map data, storing only essential information in shard files. By taking out only the necessary map elements for specific geographical sections, the system reduces data transmission requirements and processing overhead while preserving the ability to access complete information when required.
2Reliability
If monolithic map data files are used to store high-definition maps for autonomous vehicles, then complete geographical coverage and semantic object information are provided, but data upload and management become difficult
Solution Approach 1:
The patent segments monolithic map data into multiple manageable shard files organized in a hierarchical directory structure. This segmentation makes data upload, storage, and management more efficient by breaking down large files into smaller, more manageable units that can be processed and distributed more easily.
Solution Approach 2:
The patent introduces a hierarchical directory structure dimension to organize shard files, adding a spatial organization layer to the data management system. This dimensional change enables more efficient data retrieval, storage, and management by providing a structured approach to handling map data across multiple levels of organization.
3Loss of information
If large monolithic map data files are used, then comprehensive map information is available, but network capacity requirements increase
Solution Approach 1:
The patent extracts only the essential semantic objects and geographical section information needed for autonomous vehicle operation, storing them in compact shard files. By taking out only the necessary information rather than transmitting complete monolithic map data, the system significantly reduces network capacity requirements while maintaining the completeness of essential map information.
Solution Approach 2:
The patent changes the data structure parameters from monolithic format to segmented shard format, optimizing the organization and size of stored information. This parameter change enables more efficient data transmission and storage while preserving the completeness of essential map information for autonomous vehicle operations.
Data Source
AI summary
A method for storing shard data and data formats for storing shard data are proposed. Shard data entries are generated and stored, wherein each shard data entry comprises a definition of one or more semantic objects covered by the shard data entry. Shard metadata is generated and stored, wherein the metadata comprises references to the shard data entries and, for each of the shard data entries, data representative of a bounding box indicative of an area in a geographical area that is covered by the shard data entry.