Vehicle Localization via Feature Data Consolidation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle localization systems in autonomous or highly automated driving require precise data transmission and processing, which is inefficient due to repeated features in sensor data, leading to increased information volume and processing load.
Innovation Solution
The system transmits each feature only once and updates the position of recurring features, using a combination of sensor data from surround sensors and communication interfaces, with a filter to detect regularly recurring patterns, allowing for precise localization by superimposing data from different sources.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If all image features are transmitted separately to the evaluation unit, then complete information is provided for precise localization, but the quantity of data to be transmitted and processed increases significantly
Solution Approach 1:
The patent merges multiple transmissions of the same feature into a single transmission by combining feature data from different time points. The evaluation unit receives feature information once and updates it with positional changes from subsequent measurements, thereby consolidating redundant data transmissions while preserving complete localization information.
Solution Approach 2:
The system performs preliminary processing by identifying and storing feature information from initial sensor measurements before transmission. By pre-processing the data to extract only essential feature characteristics and their positions, the system reduces the overall data quantity that needs to be transmitted while maintaining the precision required for accurate localization.
2Reliability
If features occurring multiple times are transmitted each time they are detected, then up-to-date position information is provided, but the processing load and information volume increase
Solution Approach 1:
The system dynamically adjusts the transmission strategy based on feature recurrence. Instead of uniformly transmitting all detected features, the system adapts by identifying recurring features and applying optimized transmission only for their positional updates. This dynamic approach maintains data currency for all features while significantly improving processing efficiency by reducing redundant transmissions.
3Loss of energy
If data transmission from surround sensors to evaluation unit is reduced, then communication bandwidth is conserved, but localization precision may be compromised
Solution Approach 1:
The system extracts only the essential and changing information from sensor data for transmission. By separating static feature characteristics (which are transmitted once) from dynamic positional information (which is updated only when changed), the system removes redundant data transmissions while preserving all information necessary for precise localization, thereby conserving communication bandwidth without compromising precision.
Data Source
AI summary
A method and a device for localizing a vehicle in its surroundings, the vehicle having surround sensors, which at first times detect views of the surroundings using the surround sensors as sensor views and supply these to an evaluation unit, and having a communication interface, via which at second times current surroundings data regarding the current surroundings of the vehicle are transmitted to the evaluation unit, and the localization of the vehicle occurs in that in the evaluation unit the surroundings data, which were detected by the surround sensors at first times, and the temporally corresponding surrounding data, which were transmitted via the communication interface, are superimposed on one another. If it is detected that features in the surroundings data detected by the sensors and/or features in the surroundings data supplied via the communication interface occur multiple times in the data pertaining to one point in time and these represent one or multiple objects, these are transmitted only once to the evaluation device and, for a repeated occurrence of the features in the data pertaining to one point in time, only the positional data of the repeatedly occurring object are transmitted anew.