SYSTEM AND METHOD FOR ESTIMATING THE CONTINUOUS EXISTENCE OF REAL-WORLD MAP FEATURES BASED ON DISTRIBUTED VEHICLE OBSERVATIONS IN A DATA PROTECTION-PROTECTIVE WAY
Patent Information
- Application Number
- DE602022012384
- Authority / Receiving Office
- DE · DE
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2022-11-17
- Publication Date
- 2025-03-26
- Estimated Expiration
- 2042-11-17
AI Technical Summary
Existing navigation systems face challenges in maintaining accurate and up-to-date geographic map data, particularly for temporary features like construction zones or signs that change over time, as they rely on frequent observations to confirm feature existence, leading to inefficiencies and inaccuracies in route planning.
A method that estimates the likelihood of a map feature's existence using anonymized data from vehicle observations, modeling observation frequency as a Poisson process and applying Bayes' rule to determine the feature's presence, allowing for proactive removal of features from map data when likelihood falls below a threshold, ensuring map data accuracy without direct observation.
This approach enables efficient and accurate updating of geographic map data, improving navigation services by providing optimal routes that account for features' actual presence, reducing storage and energy needs, and enhancing user experience.