Campaign Management Platform for Map Data Validation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Automobile manufacturers and service providers face significant challenges in efficiently generating and maintaining high-quality digital map data for autonomous driving, particularly due to the high computational and network resource requirements of crowdsourced sensor data transmission.
Innovation Solution
A campaign management platform is introduced to coordinate map data validation, discovery, and update campaigns, which determines the minimum number of vehicles needed to collect sensor data within a data freshness threshold, optimizing data transmission and processing by targeting specific vehicles and geographic areas.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If crowdsourced sensor data is transmitted from all vehicles to generate or maintain digital map data, then map data quality is improved, but network bandwidth consumption and processing resources increase significantly
Solution Approach 1:
The patent applies local quality by differentiating between validated and unvalidated road segments. Only unvalidated segments are targeted for sensor data collection, while already validated segments are excluded. This selective approach ensures high map data quality for critical areas while minimizing unnecessary network bandwidth consumption from vehicles passing through already validated regions.
Solution Approach 2:
The patent segments the road network into validated and unvalidated road segments based on campaign results. This segmentation allows the system to treat different geographic areas differently - requesting sensor data only from vehicles traversing unvalidated segments while ignoring vehicles on validated segments, thereby optimizing the balance between map data quality and network resource utilization.
2Reliability
If sensor data is collected from a large number of vehicles to ensure data freshness, then map data up-to-date status is improved, but the number of vehicles required and processing overhead increase
Solution Approach 1:
The patent applies partial action by requesting sensor data from only the minimum necessary number of vehicles required to achieve data freshness for unvalidated segments, rather than collecting data from all vehicles. The system determines a target number of vehicles based on the specific validation needs of each campaign, avoiding excessive data collection and processing overhead while maintaining adequate data freshness.
Solution Approach 2:
The patent implements periodic action through validation campaigns that are triggered based on data freshness thresholds and confidence levels. Rather than continuously processing data from all vehicles, the system periodically initiates validation campaigns when map data approaches freshness thresholds, efficiently balancing data reliability with processing productivity by acting only when necessary.
3Measurement precision
If validation campaigns are initiated for all road segments, then comprehensive map validation is achieved, but computational resources and time consumption increase
Solution Approach 1:
The patent segments the validation process into targeted campaigns focused only on unvalidated road segments identified from previous map data analysis. Instead of initiating validation for all road segments uniformly, the system divides the road network into validated and unvalidated portions, applying validation resources only where needed, thereby achieving comprehensive validation coverage for critical areas while minimizing overall validation time and computational resource consumption.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An approach is provided for a campaign management platform to validate map data. The approach, for example, involves determining, by the campaign management platform, a road segment that has been previously mapped in a geographic database. The approach also involves generating a sensor data request specifying a sensor data collection event to be performed on the road segment. The sensor data collection event is part of a campaign to validate the digital map data for the road segment, a geographic area including the road segment, or a combination thereof. The approach further involves transmitting the sensor data request to a target number of vehicles. The target number of vehicles perform the sensor data collection event on the road segment to collect sensor data. The approach further involves processing the sensor data to validate the digital map data for the road segment.