High Definition Map Generation for Autonomous Vehicle Navigation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional maps lack the accuracy and timeliness required for safe navigation of autonomous vehicles, as they are expensive to create and maintain, and rely on outdated data, which can lead to errors in vehicle location and route determination due to limitations in sensor data and GPS accuracy.
Innovation Solution
A vehicle computing system generates high-definition maps using data from online systems, creating a connected graph of lane elements to determine potential routes with low error, allowing for accurate and up-to-date navigation by selecting the best route based on error measures and updating map data regularly.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional maps are used for autonomous vehicle navigation, then the system is simple and easy to maintain, but the location accuracy and route precision are insufficient (GPS accuracy of 3-5 meters with large error conditions exceeding 100 meters)
Solution Approach 1:
The patent segments the map data into a hierarchical structure with a low-resolution base map providing general geographic information and a high-resolution overlay layer containing precise lane geometry, traffic fixtures, and navigation-relevant details. This segmentation allows the system to achieve centimeter-level accuracy (improving measurement precision) while keeping the base map simple and the complex high-resolution data confined to specific overlay structures (managing device complexity).
Solution Approach 2:
The patent adds a dimensional layer by creating a multi-layered map structure where a two-dimensional base map is enhanced with a third dimension of detail through overlay layers. The base map provides coarse geographic context while overlay layers add precise lane-level geometry and semantic information, enabling accurate vehicle localization without requiring the entire map to be high-resolution (balancing measurement precision with device complexity).
2Reliability
If survey teams with expensive high resolution sensors are used to create maps, then the map accuracy is high, but the cost and time consumption are excessive (taking possibly months to complete a map and requiring a thousand cars to keep up-to-date)
Solution Approach 1:
The patent enables autonomous vehicles to self-generate high-resolution map data using their own sensors while navigating. Vehicles continuously collect and contribute lane geometry, traffic fixture, and road condition data to a centralized system, which processes and distributes updates to the fleet. This self-service approach eliminates the need for expensive survey teams (improving productivity) while maintaining high map accuracy through aggregated real-world sensor data (presving reliability).
Solution Approach 2:
The patent transforms map creation from a periodic, resource-intensive survey process into a continuous data collection process. As autonomous vehicles continuously navigate and sense their environment, map data is continuously updated and refined. This continuous action ensures maps remain current with road changes without requiring repeated expensive survey expeditions (improving productivity while maintaining reliability).
3Loss of information
If conventional maps are updated frequently to capture road changes (5-10% per year), then the data freshness is improved, but the cost and resource requirements increase significantly (a thousand survey cars would be needed)
Solution Approach 1:
The patent merges the map updating function with the autonomous vehicle's primary navigation function. Instead of deploying dedicated survey vehicles, the system combines data collection from vehicles already performing navigation tasks. Multiple vehicles' sensor data is merged and processed to create updated map versions, achieving frequent updates (reducing information loss) without requiring additional survey fleet resources (reducing quantity of substance).
Solution Approach 2:
The patent makes autonomous vehicles multi-functional by enabling them to simultaneously perform navigation and map data collection. The same sensors used for vehicle localization and obstacle detection are also used to capture and contribute to map updates. This universality allows the system to maintain fresh map data (reducing information loss) without requiring separate survey vehicles (reducing quantity of substance).
Data Source
AI summary
A system generates a high definition map for an autonomous vehicle to travel from a source location to a destination location. The system determines a low resolution route and receives high definition map data for a set of geographical regions overlaying the low resolution route. The system uses lane elements within the geographical regions to form a set of potential partial routes. The system calculates the error between the potential partial route and the low resolution route and removes potential partial routes with errors above the threshold. Once completed, the system selects a final route and sends signals to the controls of the autonomous vehicle to follow the final route. The system determines whether surface areas adjacent to a lane that are not part of the road are safe for the vehicle to drive in case of emergency. The system stores information describing navigable surface areas with representations of lanes.