HD Map Change Updates for Low-Latency Autonomous Driving
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for updating high-definition maps in autonomous vehicles are inefficient and resource-intensive, as they require frequent surveys by high-quality vehicles to reflect reality changes, leading to significant delays and resource overheads in distributing map updates.
Innovation Solution
A server system generates and distributes map change data independently from HD map data, using observational data from various sources to identify and update map features, along with updated metadata to reflect confidence levels, allowing vehicles to efficiently apply these changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frequent surveys by high-quality vehicles are conducted to update HD maps, then map accuracy is improved, but resource consumption and time delay increase
Solution Approach 1:
The patent segments map updates into two independent components: full HD map data and incremental map change data. Map change data captures only modifications (additions, deletions, updates) since the last full map version. This segmentation allows vehicles to receive lightweight update packets instead of complete maps, reducing transmission time and processing overhead while maintaining accuracy.
Solution Approach 2:
The system performs preliminary processing of observational data to identify changes before distribution. Survey vehicles pre-process their observational data to generate map change data, identifying what has changed before transmitting to the server. This preliminary action filters out redundant data and prepares update packages in advance, reducing actual update delay.
2Loss of time
If full HD map data is redistributed frequently, then map updates are timely, but data transmission overhead increases
Solution Approach 1:
The patent extracts only the changed portions from complete map data. The map change data structure contains only modifications (new features, modified features, deleted features) with their identifiers, types, and change descriptions. This extraction eliminates transmission of unchanged map content, reducing data volume by orders of magnitude while maintaining update timeliness.
Solution Approach 2:
The system uses a copy-based update mechanism where map change data is essentially a differential copy of the full map. Instead of transmitting complete map copies, the system transmits compact change records that can be applied to existing map versions. This copying approach minimizes transmission overhead while ensuring vehicles receive timely updates.
3Loss of information
If complete HD map data is processed by vehicles, then comprehensive map information is available, but processing resources are consumed
Solution Approach 1:
Vehicles perform partial processing by applying only the necessary map changes to their existing map versions. Instead of processing complete map data, vehicles selectively apply changes to specific map features identified in the map change data. This partial action approach maintains information completeness for relevant features while dramatically reducing computational energy consumption.
Data Source
AI summary
Disclosed herein is a technique for processing map change data in a map client that uses the map changes to apply changes and updates to current map data. The map client provides the updated map data to other components in an autonomous vehicle. The map change data describes the processing in the map client to obtain updated map data from the map data.