Vehicle Landmark Identification via Backend Image Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for highly accurate vehicle position determination, such as differential GPS and LIDAR, are costly and complex, and require multiple landmarks for reliable position finding in automated driving applications.
Innovation Solution
A method utilizing a vehicle's image recording and evaluation devices to capture and transmit images to a data processing station for landmark identification, where image analysis determines suitable landmarks, reducing data transmission frequency and leveraging available computation power for accurate position determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If differential GPS or LIDAR algorithms are used for precise vehicle position determination, then measurement precision is improved, but device complexity and cost increase
Solution Approach 1:
The patent creates a digital copy of the physical environment by capturing images and identifying landmarks within them. These landmark positions and visual features are stored as digital data that can be used for position determination, replacing the need for expensive physical sensors like LIDAR while maintaining measurement precision through image-based landmark recognition
Solution Approach 2:
The patent replaces mechanical/optical measurement systems (differential GPS receivers and LIDAR sensors) with an image processing system. Instead of using complex hardware to directly measure position, the system uses cameras to capture images and algorithms to identify landmarks, substituting mechanical measurement with optical capture and computational analysis
2Measurement precision
If multiple landmarks are captured and made available in the vehicle for evaluation, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent segments the complex task of position determination into two parts: image capture and landmark identification. The vehicle's image recording device captures images, while the backend server performs landmark identification and position calculation. This segmentation allows the vehicle to use simpler hardware while the complex processing is distributed to the server
Solution Approach 2:
The patent introduces a backend server as an intermediary between the vehicle's image capture device and the position determination function. The server receives images, identifies landmarks, and provides position information to the vehicle, acting as a mediator that handles the complex landmark processing while the vehicle maintains simpler onboard systems
3Measurement precision
If images are recorded and evaluated at every vehicle position, then measurement precision is improved, but loss of energy increases due to frequent data transmission
Solution Approach 1:
The patent implements periodic action by having the vehicle transmit images to the backend server at specific intervals or under specific conditions rather than continuously. The server processes these periodic transmissions to identify landmarks and update position information, reducing the frequency of data transmission while maintaining adequate position determination accuracy
Data Source
AI summary
A vehicle takes and evaluates an image of the surroundings of the vehicle. If the result of the evaluation is that the image possibly contains a static object that could be suitable as a landmark, then the image and a position of the vehicle ascertained by the vehicle at the location at which the image is taken are transmitted to a data processing station. Image analysis of the received image allows the data processing station to establish whether the supposedly static object is suitable as a landmark. When the object has been verified as a landmark, the data processing station creates a data record describing the object, which data record is transmitted to the vehicle.