Onboard HD Map Generation From SD Maps for Lane-Level Routing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing autonomous vehicle systems face challenges in generating and maintaining high-definition (HD) maps at scale, particularly in dynamic environments, requiring continuous updates and human validation, which are costly and inefficient.
Innovation Solution
Utilizing real-time perception data from vehicle sensors and sparse, lightweight standard definition (SD) maps to generate HD maps and lane-level trajectories onboard, leveraging machine learning and neural networks to enhance online road estimation and trajectory planning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If HD maps are generated and maintained using traditional offline methods with human validation, then map accuracy is improved, but cost and time consumption increase significantly
Solution Approach 1:
The system enables autonomous vehicles to self-generate and self-update HD maps using their own sensor data and onboard processors, eliminating the need for centralized offline processing and human validation. The vehicle independently performs perception data generation, map updating, and quality assessment
Solution Approach 2:
The patent replaces manual human validation processes with automated machine learning algorithms and neural networks that can autonomously validate and integrate map data. The system substitutes mechanical offline processing with online computational methods using perception data
2Reliability
If HD maps are updated continuously to reflect dynamic environments, then navigation reliability is improved, but system complexity and computational resources increase
Solution Approach 1:
The system implements dynamic map updating where HD maps are continuously adapted to reflect changes in the environment. The map data structure supports real-time modifications based on current perception data, allowing the system to respond to dynamic conditions such as construction zones, temporary road closures, and changing traffic patterns
Solution Approach 2:
The patent segments the map updating process into modular components: perception data acquisition, feature extraction, map integration, and validation. This modular approach reduces system complexity by allowing each component to be independently optimized and managed
3Measurement precision
If HD maps are generated offline and pre-loaded onto vehicles, then initial map quality is improved, but adaptability to new environments deteriorates
Solution Approach 1:
The system performs preliminary actions by pre-loading SD maps and base HD map structures onto vehicles before deployment. These pre-loaded maps provide a foundation that is then enhanced in real-time through online perception data, combining the benefits of both offline preparation and online adaptation
Solution Approach 2:
The patent implements parameter changes by transitioning from static, pre-loaded map data to dynamic, real-time updated map data. The system modifies map parameters such as lane geometry, traffic signs, and road conditions based on current sensor inputs, enabling continuous adaptation to new environments while maintaining data quality
Data Source
AI summary
Methods and systems for generating a HD map and lane trajectory for an autonomous vehicle based on an SD map. Images from one or more image sensors mounted on a vehicle are received. Via a vehicle processor, perception data is generated based on the received images, wherein the perception data provides a representation of an environment proximate to the vehicle. A standard definition (SD) map corresponding with the environment proximate to the vehicle. The vehicle processor generates a high definition (HD) map corresponding with the environment proximate to the vehicle based on the SD map and the perception data. The vehicle processor also generates lane-level trajectory associated with a planned route for the vehicle utilizing the HD map.