HAV Localization Map Filtering for Hidden Landmark Exclusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for localizing highly automated vehicles in digital maps face challenges due to hidden landmarks, leading to unnecessary data transmission and poor localization accuracy, which compromises system reliability.
Innovation Solution
A method that senses features of semi-static objects using on-board sensors, classifies them, and creates a local driving-environment model excluding hidden landmarks, thereby transmitting only relevant landmarks for localization, enhancing robustness and accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all landmarks are transmitted to the HAV for localization, then localization accuracy may be maintained, but data transmission volume increases and computing complexity increases
Solution Approach 1:
The patent extracts only the necessary subset of landmarks from the complete set. The server identifies and transmits only those landmarks that are currently visible and relevant for localization, excluding hidden or irrelevant landmarks. This extraction principle reduces data transmission volume while maintaining localization accuracy by providing only the essential information needed.
2Loss of information
If all landmarks are transmitted to the HAV, then complete information is available for matching, but computing complexity in the vehicle increases
Solution Approach 1:
The server performs the extraction of relevant landmarks before transmission, moving the computational burden from the vehicle to the server. By pre-processing and filtering landmarks on the server side based on visibility and relevance, the vehicle receives a reduced set of landmarks that requires less computing power for matching and localization.
Solution Approach 2:
The server acts as an intermediary between the complete landmark database and the HAV. It processes the landmark information, filters out irrelevant or hidden landmarks, and transmits only the necessary subset to the vehicle. This intermediary processing reduces the computing complexity in the vehicle while maintaining information completeness for localization.
3Loss of information
If hidden landmarks are included in transmission, then data completeness is maintained, but unnecessary data transmission occurs and localization accuracy deteriorates
Solution Approach 1:
The server performs preliminary analysis of landmark visibility and relevance before transmission. By pre-identifying which landmarks are currently visible and relevant based on the HAV's position and trajectory, the system avoids transmitting hidden or irrelevant landmarks. This preliminary filtering action ensures that only useful data is transmitted, improving both reliability and efficiency.
Data Source
AI summary
A method is described for localizing a highly automated vehicle (HAV) in a digital localization map, including the following steps: S1 sensing features of semi-static objects in an environment of the HAV with the aid of at least one first sensor; S2 transmitting the features of the semi-static objects as well as the vehicle position to an evaluation unit; S3 classifying the semi-static objects, the feature “semi-static” being assigned to the semi-static objects as a result of the classification; S4 transferring the features of the semi-static objects into a local driving-environment model of the HAV, when creating the local driving-environment model, it being checked whether landmarks suitable for localizing the HAV are hidden by the semi-static objects in terms of the position and/or an approach trajectory of the HAV; S5 transmitting the local driving-environment model to the HAV in the form of a digital localization map, the digital localization map containing only those landmarks suitable for localizing the HAV which are not hidden by semi-static objects in terms of the position and/or an approach trajectory of the HAV; and S6 localizing the HAV using the digital localization map. In addition, a corresponding system and computer program are described.