Vehicle Surroundings Model Verification Using Sensor Overlap
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle surroundings models are prone to errors and inaccuracies due to the absence of effective reference checks for sensor-based measured data, which hinders precise long-term driving maneuvers.
Innovation Solution
A method involving multiple sensor sets with overlapping scanning areas to create and combine surroundings models, allowing for redundancy and verification, thereby reducing inaccuracies and enhancing model accuracy through data averaging and correction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If sensor-based measured data is used to create surroundings models, then the implementation of automated driving maneuvers is enabled, but the models are subject to errors and inaccuracies due to absence of reference checks
Solution Approach 1:
The patent implements feedback by comparing surroundings models from multiple sensor sets and using the overlap area as a reference to verify and correct measured data. The system continuously checks the consistency of detections across different sensors and adjusts the surroundings model based on discrepancies found in the overlapping region, thereby improving measurement precision while maintaining automated driving capabilities.
Solution Approach 2:
The overlap area between scanning areas of different sensor sets serves as an intermediary reference region. This intermediary zone allows indirect verification of measured data by comparing detections from multiple sensors without requiring external reference objects, thus resolving the accuracy issue while preserving automation.
2Measurement precision
If multiple sensor sets with overlapping scanning areas are used, then surroundings models can be verified and optimized, but the system complexity and data processing requirements increase
Solution Approach 1:
The patent segments the verification process by dividing the environment into overlapping scanning areas from multiple sensor sets. Each sensor set independently creates a surroundings model, and the system then compares these segmented models in the overlap region. This segmentation approach allows verification without requiring a completely different system architecture, thus managing complexity while improving precision.
3Reliability
If redundancy is created in the overlap area by combining multiple sensor sets, then measured data can be checked and optimized, but the quantity of data to be processed increases
Solution Approach 1:
The patent extracts only the critical verification information from the redundant data in the overlap area. Instead of processing all redundant data equally, the system identifies and focuses on discrepancies between sensor models in the overlap region, extracting only the necessary correction information. This reduces the effective data volume to be processed while maintaining reliable verification capabilities.
Data Source
AI summary
A method for optimizing a surroundings model by at least one control unit, measured data being received from a first sensor set and at least one second sensor set. The first sensor set includes a first scanning area, and the second sensor set includes a second scanning area, the first scanning area and the second scanning area partially overlapping in an overlap area. A surroundings model is created for each sensor set based on the received measured data of the particular sensor set. The at least two surroundings models are compared to one another based on the overlap area and being verified. The at least two surroundings models are combined into an optimized surroundings model. A system, a control unit, a computer program, and a machine-readable memory medium, are also described.