Ground Vehicle Localization Using Aerial Data Layers and Road Alignment
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current localization solutions for autonomous vehicles rely on high-definition maps that require driving by multiple vehicles and depend on predetermined landmarks, leading to inefficiencies and inaccuracies in vehicle location determination.
Innovation Solution
A method utilizing a combination of air-based data, environmental information, and database-populated data layers to enhance vehicle localization, incorporating road object location information and real-time ground perception data for improved accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If high-definition maps with predetermined landmarks are used for localization, then localization reliability is improved, but device complexity and time consumption increase due to requiring multiple vehicles to drive and update maps
Solution Approach 1:
The patent extracts the essential localization function from the complex high-definition map system. Instead of relying on comprehensive maps with predetermined landmarks that require multiple vehicles to update, the invention uses a simplified approach with road object location information and aerial images, removing the unnecessary complexity while maintaining localization reliability
Solution Approach 2:
The patent segments the localization problem into two independent components: obtaining road object location information from a database and capturing real-time aerial images of road objects. This segmentation eliminates the need for a complex integrated high-definition map system, allowing each component to function independently and reducing overall system complexity
2Measurement precision
If high-definition maps requiring multiple vehicle drives are used, then localization accuracy is improved, but productivity decreases due to time-consuming map updates
Solution Approach 1:
The patent applies preliminary action by pre-populating a database with road object location information from aerial images before vehicles need localization. This eliminates the need for vehicles to spend time updating maps during operation, maintaining high localization accuracy while significantly improving vehicle productivity and operational efficiency
Solution Approach 2:
The patent creates a simplified copy of the high-definition map concept by using a database of road object locations combined with real-time aerial images, rather than using full high-definition maps. This copying approach achieves comparable localization accuracy without the time-consuming update requirements, thereby improving vehicle productivity
3Reliability
If predetermined landmarks are required for localization, then localization reliability is improved, but adaptability worsens due to inability to locate in areas without landmarks
Solution Approach 1:
The patent makes the localization system universal by using road objects (such as road signs, barriers, and markings) that exist in diverse environments rather than relying on specific predetermined landmarks. This allows the same localization approach to work reliably across different geographic locations and road types, significantly improving adaptability while maintaining reliability
Data Source
AI summary
A method that includes obtaining, by a processor associated with the vehicle, a cross-view based localization of the vehicle that is determined by using air based data in accordance with environmental information sensed by a sensor of the vehicle at the region of the vehicle. Then, obtaining, by accessing a database that is populated to contain a data layer, data layer information regarding locations of a given road setting within the region of the vehicle; obtaining, in real time, ground detection output that is being generated for the given road setting by a perception unit of the vehicle; and providing real-time fine-tuned localization of the vehicle, by continuous alignment of the ground detection output in accordance with the data layer information, for the given road setting.