Detailed Map Format for Autonomous Vehicle Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing virtual maps lack sufficient accuracy and detail to support optimized autonomous vehicle operation, which is crucial for precise control and localization of self-driving systems.
Innovation Solution
A detailed map format that includes lane segments formed of waypoints, border segments formed of borderpoints, and associated information such as border type and color, allowing for precise control and localization of autonomous vehicles by determining distances and implementing driving maneuvers based on these features.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing virtual maps are used, then the system is simple to operate, but the localization accuracy and control precision are insufficient for optimized autonomous operation
Solution Approach 1:
The map is segmented into multiple hierarchical levels: overview maps providing general road layouts and detailed maps providing precise lane-level information. This segmentation allows the system to use simple overview maps for general navigation while switching to complex detailed maps only when high precision localization and control are required, thus improving measurement precision without permanently increasing device complexity.
Solution Approach 2:
Different regions of the map have different levels of detail and quality. Areas requiring high precision autonomous operation have detailed lane markings, border segments, and geographic features, while other areas use simplified representations. This local quality approach ensures high localization accuracy is available where needed without making the entire map system unnecessarily complex.
2Manufacturing precision
If detailed map format with border segments is used, then the control precision and lane positioning accuracy improve, but the data processing complexity and computational requirements increase
Solution Approach 1:
Border segments, lane segments, and geographic features are pre-processed and stored in the detailed map format before autonomous operation. This preliminary action organizes complex spatial data into structured formats with defined relationships, reducing the computational burden during real-time operation and allowing high precision lane positioning without excessive data processing complexity.
Solution Approach 2:
The detailed map format acts as an intermediary layer between raw sensor data and vehicle control systems. It pre-integrates border segments, lane segments, and geographic features into a unified coordinate system, mediating the complexity by providing a structured representation that simplifies downstream processing while maintaining high positioning accuracy.
3Reliability
If high detail map format is implemented, then the autonomous vehicle control and localization performance improve, but the memory requirements and storage capacity needed increase
Solution Approach 1:
The map data is segmented into overview maps and detailed maps, allowing the system to store large volumes of detailed information in a modular fashion. Only the detailed map portions relevant to the current operating context are actively used, reducing the effective memory burden while maintaining high reliability where needed.
Solution Approach 2:
High-detail map data with precise border segments and geographic features is stored only for regions where autonomous operation is currently active or anticipated. Other regions use simplified representations, reducing overall data volume while maintaining reliability in critical operational zones.
Data Source
AI summary
A computer-readable detailed map format is disclosed. The detailed map format includes a lane segment and one or more border segments. The map format can be used in the operation of an autonomous vehicle. A current location of the autonomous vehicle can be determined. The current location of the autonomous vehicle can be compared to the computer readable map format. A distance between the current location of the autonomous vehicle and an edge of the lane segment at a location along the lane segment can be determined by, for example, measuring a distance between the current location of the autonomous vehicle and a portion of the border segment closest to the current location of the autonomous vehicle. A driving maneuver can be determined based at least in part on the determined distance. One or more vehicle systems of the autonomous vehicle can be caused to implement the determined driving maneuver.