Digital Map Generation Using Geometric Image Data for Vehicle Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital maps used for vehicle navigation require high accuracy in indicating the positions of objects, such as lane boundaries and landmarks, and need to be updated frequently to account for changes in the environment.
Innovation Solution
A system and method for producing a digital map from data affiliated with images of a location, excluding pixel color data, which includes information about the pose of a camera and the positions of points on lane boundaries, road boundaries, or landmarks, allowing for the creation of high-definition maps that can be transmitted to vehicles for navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If pixel color data is included in image data for digital map production, then semantic information about objects can be captured, but data transmission time and processing complexity increase
Solution Approach 1:
The patent extracts and transmits only the essential geometric and positional data (lane boundaries, road boundaries, landmarks) from image data, excluding redundant pixel color information. This extraction principle reduces data transmission time while maintaining the necessary semantic information for navigation and vehicle control.
2Measurement precision
If high-definition maps with detailed object positions are produced, then vehicle control accuracy improves to decimeter level, but data processing and map update complexity increase
Solution Approach 1:
The patent segments the digital map into distinct geometric components (lane boundaries, road boundaries, landmarks) and processes each segment independently. This segmentation approach enables precise positioning at decimeter level while reducing overall processing complexity through modular data handling and structured organization of map features.
3Reliability
If frequent map updates are performed to reflect environmental changes, then navigation reliability improves, but data processing time and system resource consumption increase
Solution Approach 1:
The patent implements periodic map updates based on detected environmental changes, using structured geometric data that enables efficient comparison and validation. This periodic update mechanism maintains high navigation reliability by reflecting current road conditions while optimizing resource consumption through change-detection-based triggering and efficient data structures.
Data Source
AI summary
A system for producing, from data affiliated with images of a location, a digital map can include a processor and a memory. The memory can store a productions module and a communications module. The production module can include instructions that cause the processor to produce, from the data affiliated with the images of the location, the digital map. The data, for an image, can exclude pixel color data, but can include information about: (1) a pose of a camera that produced the image and (2) one or more of a position of a point on: (a) a lane boundary of a lane of a road in the image, (b) a road boundary of the road, or (c) a landmark in the image. The communications module can include instructions that cause the processor to transmit the digital map to a vehicle to be used to control a movement of the vehicle.