Environmental Model Update Validation via Sensor Quality Plausibility
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current environmental models for autonomous vehicles are prone to incorrect changes due to outdated or imprecise sensor data, which can compromise safety and reliability in fully automated driving systems.
Innovation Solution
A method is implemented where a central server performs a plausibility check on detected changes in the environmental model using data from vehicles with higher sensor quality, ensuring that only reliable updates are made to the digital map, and optimizing sensor data processing in vehicles with lower sensor quality to prevent incorrect change recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If sensor data from vehicles with lower sensor quality is used to update the environmental model, then the system can utilize more data sources and improve coverage, but incorrect changes may be introduced compromising safety and reliability
Solution Approach 1:
The patent applies local quality by differentiating the treatment of sensor data based on the quality level of individual vehicles. High-quality vehicle data is accepted directly for environmental model updates, while low-quality vehicle data undergoes additional plausibility checks before acceptance. This localized differentiation ensures that data from lower-quality sources does not compromise overall system reliability while still utilizing the broader data coverage.
Solution Approach 2:
The patent implements feedback mechanisms where the central server performs plausibility checks on detected changes using reference data, and vehicles receive feedback about the validity of their detected changes. This feedback loop allows the system to learn from previous incorrect updates and improve future data acceptance decisions, thereby maintaining reliability while incorporating data from multiple sources.
2Reliability
If plausibility checks are performed on all detected changes using reference data from high-quality vehicles, then the reliability and safety of environmental model updates are improved, but the processing time and system complexity increase
Solution Approach 1:
The patent reduces system complexity by applying plausibility checks selectively rather than universally. The central server identifies which detected changes require validation based on the quality level of the reporting vehicle, performing comprehensive checks only for low-quality data sources while accepting high-quality data without additional validation. This localized approach maintains safety while minimizing processing overhead.
Solution Approach 2:
The patent applies partial action by performing plausibility checks only when necessary, rather than on all detected changes. The system uses reference data from high-quality vehicles to validate changes reported by low-quality vehicles, but does not subject all data to the same level of scrutiny. This partial validation approach ensures safety where needed while avoiding unnecessary processing complexity.
3Speed
If the system updates the environmental model immediately upon detecting changes, then the responsiveness and up-to-date status of the model is improved, but incorrect updates may be propagated quickly compromising safety
Solution Approach 1:
The patent resolves this contradiction by differentiating update processing based on data quality. Changes detected by high-quality vehicles are implemented immediately, maintaining fast response time. Changes from low-quality vehicles undergo plausibility checks before implementation, preventing incorrect updates while maintaining overall system responsiveness through selective immediate updates.
Solution Approach 2:
The patent applies preliminary action by performing plausibility checks on detected changes before implementing them in the environmental model. The central server validates changes using reference data from high-quality vehicles before accepting them, ensuring accuracy is maintained while still allowing timely updates through the efficient processing of validated changes.
Data Source
AI summary
A method is disclosed for checking detected changes to an environmental model of a digital area map. According to the method, a first vehicle with one or more sensors detects an environmental change from an environmental model of a digital area map. Data are transferred from the first vehicle to a central server, wherein the data include information on the detected environmental change, position data of the detected environmental change, as well as information using which the sensor quality of the one or more sensors can be detected. Depending on the sensor quality of the one or more sensors, the central server performs a plausibility check of the detected environmental change using data that are detected by one or more sensors of an additional vehicle.