Autonomous Driving Map Fusion for Inconsistent Recording Targets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous driving systems face challenges in acquiring necessary information for navigation when using maps with different recording targets, as each map is created according to unique standards, leading to inconsistencies in content and methods for data acquisition.
Innovation Solution
An autonomous driving device that includes multiple maps with different recording targets, using a control unit to determine necessary information using a unified method by treating absent content as a null value, allowing the device to perform autonomous driving across various maps with consistent methodology.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If multiple maps with different recording targets are used, then the coverage and adaptability of the autonomous driving system is improved, but the complexity of acquiring and processing information from different map standards increases
Solution Approach 1:
The patent introduces a standardization layer that acts as an intermediary between maps with different recording targets and the autonomous driving control system. This layer converts various map data formats and content structures into a unified standard format, enabling the system to process information from multiple map sources without requiring separate processing methods for each map type.
Solution Approach 2:
The patent creates a universal information acquisition method that can handle multiple map types with different recording targets through a single standardized interface. The control unit uses the same processing logic regardless of whether the map records detailed road geometry, basic navigation data, or specialized information,从而实现 one system serving multiple map standards.
2Manufacturing precision
If detailed maps are created for all roads, then the accuracy and completeness of autonomous driving information is improved, but the cost of map creation and maintenance increases significantly
Solution Approach 1:
The patent applies local quality by allowing different map regions to have different levels of detail and recording targets based on local requirements. Urban areas with complex intersections may use maps with detailed recording targets, while rural areas may use maps with basic navigation information. The standardization layer adapts to each region's specific needs without requiring uniform high-detail mapping everywhere.
3Adaptability or versatility
If maps are created by different companies or organizations according to their own standards, then the diversity and availability of map sources is improved, but the consistency and uniformity of information acquisition methods deteriorates
Solution Approach 1:
The patent changes the parameter of information representation by introducing a standardized data structure and format that all map sources must conform to. The control unit acquires information using consistent methods because all maps are transformed into the same parameter format, regardless of their original creation standards. This parameter standardization enables uniform information acquisition across diverse map sources.
Data Source
AI summary
An autonomous driving device includes a first map including a first content and a second content associated with positions respectively, a second map including the first content associated with the position and in which the second content is not a recording target, and a control unit performing the autonomous driving. When the autonomous driving is performed using the first map, the control unit determines information necessary for the autonomous driving using a first method based on the first content and the second content. When the autonomous driving is performed using the second map, the control unit determines the information necessary for the autonomous driving using a method same as the first method based on the first content corresponding to a second position recorded in the second map and information indicating that the second content corresponding to the second position recorded in the second map is not present.