Digital Map Verification Using Vehicle Sensor Feature Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Highly automated vehicles face challenges in accurately updating their digital maps due to short-term changes like construction sites or accidents, which can lead to safety issues and high costs associated with traditional mapping methods, and existing sensor limitations affect the robustness of map validation.
Innovation Solution
A method for verifying a digital map by determining a current reference position, comparing actual feature properties with target properties using a variety of information sources, and classifying the map as up-to-date or not based on deviation thresholds, with the option to request updates or dispatch mapping vehicles when necessary.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional mapping methods (motor vehicles, airplanes, satellites) are used to update digital maps, then mapping completeness can be improved, but mapping costs increase significantly
Solution Approach 1:
The system uses the highly automated vehicle's own sensor data to verify and update the digital map, eliminating the need for separate expensive mapping missions. The vehicle serves dual purposes: transportation and map verification, with the object detection system continuously comparing detected features against the digital map during normal operation.
Solution Approach 2:
The system establishes a feedback loop where sensor data from the vehicle is continuously compared with the digital map, and discrepancies trigger automated update requests. This closed-loop verification process ensures map accuracy while using the vehicle's operational data rather than requiring separate mapping expeditions.
2Reliability
If frequent digital map updates are performed to capture short-term changes (construction sites, accidents), then digital map topicality is improved, but system complexity and response time requirements increase
Solution Approach 1:
Instead of continuously updating the entire digital map, the system performs partial updates only for specific map sections where discrepancies are detected. The verification is targeted to affected areas identified by the object detection system, reducing unnecessary processing and system complexity while maintaining topicality where needed.
Solution Approach 2:
The digital map verification and update process is segmented into discrete sections rather than treating the entire map as a single unit. When a discrepancy is detected in one area, only that specific section is flagged for update, allowing the rest of the map to remain unchanged and reducing overall system complexity.
3Measurement precision
If multiple sensor types are used to verify digital map, then measurement robustness is improved, but sensor-specific limitations (occlusions, lighting effects) still affect detection accuracy
Solution Approach 1:
The system merges data from multiple sensor types (cameras, radar, LIDAR, ultrasonic sensors) to verify digital map features. By combining the strengths of different sensors, the system compensates for individual sensor limitations - for example, using radar to detect features when cameras are obscured by lighting conditions or physical occlusions.
Solution Approach 2:
The system uses the digital map itself as an intermediary to cross-validate sensor detections. When sensor data is ambiguous or affected by limitations, the system refers to the digital map's stored feature information to resolve uncertainties, and vice versa, creating a mutual verification mechanism that overcomes individual sensor weaknesses.
Data Source
Figure 1
Figure 2
AI summary
The invention relates to a method for verifying a digital map in a more highly automated vehicle, in particular a highly automated vehicle, comprising the steps of: S1 providing a digital map, preferably a highly accurate digital map; S2 determining a current reference position and locating the reference position in the digital map; S3 determining at least one actual feature property of a feature in the environment of the reference position, wherein the determination is carried out using at least one information source; S4 comparing the actual feature property with a desired feature property of the feature and determining at least one difference value as the result of the comparison. The invention also relates to a corresponding device and to a computer program.