Trajectory-Based Lane Configuration for HD Map Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous vehicles lack sufficient map information for navigation due to the high cost and unavailability of high-definition (HD) maps, which typically include lane configuration details, and rely on standard navigation maps that do not provide enough detail for autonomous driving.
Innovation Solution
Derive lane configuration from analyzing the trajectories of multiple vehicles over time, using perception data from sensors to determine driving behaviors and update standard navigation maps with lane configuration information, creating a higher definition map that includes lane details such as number, width, and boundaries, which can be used for autonomous driving.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If standard navigation maps are used for autonomous vehicle navigation, then the system is simple and cost-effective, but the map information is insufficient for autonomous driving
Solution Approach 1:
The system uses autonomous vehicles themselves to collect trajectory data and generate HD map information. The vehicles' own navigation and sensor data are utilized to build the detailed maps they need, eliminating the need for separate manual mapping operations.
Solution Approach 2:
The system continuously collects trajectory data from autonomous vehicles and uses this feedback to iteratively improve and update the HD map. The map is dynamically refined based on real-world driving data, ensuring accuracy while maintaining system simplicity.
2Loss of information
If high-definition maps are generated and maintained traditionally, then the map includes detailed lane configuration information, but the cost and time for maintenance are high
Solution Approach 1:
The HD map is made dynamic rather than static. It continuously updates using real-time trajectory data from autonomous vehicles, allowing the map to adapt to changes in road conditions, new lanes, and traffic patterns without manual intervention.
Solution Approach 2:
The system performs continuous map maintenance through ongoing data collection and processing. Instead of periodic manual updates, the map is constantly refined using the continuous stream of trajectory data from vehicles in operation.
3Loss of information
If high-definition maps are generated and maintained traditionally, then the map includes detailed lane configuration information, but the generation and maintenance cost is expensive
Solution Approach 1:
The autonomous vehicles perform multiple functions: they navigate using the map, collect trajectory data, and contribute to map generation. This multi-functionality eliminates the need for dedicated mapping operations, reducing overall system cost.
Solution Approach 2:
The system generates its own HD map using data from its operational vehicles. Instead of requiring external mapping services or manual surveying, the autonomous fleet collectively builds and maintains the detailed lane configuration information it needs.
Data Source
AI summary
According to one embodiment, perception data describing a set of trajectories driven by a number of vehicles is received at a server from the vehicles or data collection agents over a network. The vehicles were driving through a road segment of a road over a period of time and their driving trajectories were captured. A trajectory analysis is the performed on the perception data using a set of rules to determine driving behaviors of the corresponding vehicles. A lane configuration of the road segment is then determined based on the driving behaviors. A map segment of a navigation map is then updated based on the lane configuration of one or more lanes within the road segment. A higher definition map can be generated based on the updates of the navigation map and the lane configuration, which can be utilized to autonomously drive an ADV subsequently.