Virtual Lane Generation for Lane Recognition Limit Situations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Lane recognition systems in vehicles fail in challenging environments, leading to difficulties in maintaining safe driving paths, especially when lane markers are obscured or complex road structures are encountered, affecting lane change and following controls.
Innovation Solution
A method and system for generating a virtual lane by processing sensor data and previous lane information, using techniques like Dead Reckoning, Extended Kalman Filter, and Weighted Least Squares with Tikhonov regularization to optimize lane coefficients and generate a virtual lane when lane recognition is limited.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If lane recognition is performed using image sensors in challenging environments (bad weather, road wear, damaged lane markers), then the system can provide lane change and following control functions, but the lane detection accuracy deteriorates and may fail completely
Solution Approach 1:
The system performs preliminary actions by predicting future lane marker positions and vehicle positions before actual lane recognition is needed. Using dead reckoning, the system calculates expected lane marker locations based on previously detected lanes and vehicle motion, preparing virtual lane information in advance that can be immediately utilized when real-time detection fails.
Solution Approach 2:
The patent introduces virtual lane markers as an intermediary element between the vehicle and the actual physical lane markers. When physical lane markers cannot be detected reliably, the virtual lane markers serve as a mediator to provide continuous lane guidance. These virtual markers are generated through coordinate transformations and predictions, bridging the gap when direct observation fails.
2Reliability
If the system uses only image sensor data for lane recognition, then the system complexity remains low, but the system fails to maintain lane tracking when lane markers are obscured or damaged
Solution Approach 1:
The system merges multiple information sources including previously detected lane markers, vehicle position data from dead reckoning, and predicted lane marker positions. By combining these elements into a unified virtual lane model, the system maintains lane tracking continuity without requiring additional complex hardware, utilizing existing data in new ways.
Solution Approach 2:
The system performs preliminary calculations of virtual lane markers using dead reckoning and coordinate transformations before actual lane recognition is needed. This advance preparation ensures that lane tracking information is available even when real-time detection fails, maintaining continuity without adding complex real-time processing requirements.
3Measurement precision
If virtual lane generation is performed using complex optimization methods (Weighted Least Squares with Tikhonov regularization), then the accuracy of virtual lane parameters improves, but the computational complexity and processing time increase
Solution Approach 1:
The system changes the parameter representation of lane markers from simple geometric descriptions to polynomial curve representations. This parameter transformation allows the use of Weighted Least Squares with Tikhonov regularization to fit smooth curves through predicted points, improving accuracy while the mathematical framework provides efficient computational solutions through closed-form or iterative optimization methods.
Solution Approach 2:
The patent replaces direct geometric interpolation methods with mathematical optimization approaches (Weighted Least Squares and Tikhonov regularization). This substitution uses algebraic and numerical methods instead of geometric constructions, providing more accurate parameter estimation through systematic optimization while maintaining computational efficiency through established mathematical algorithms.
Data Source
AI summary
A method and a system for generating a virtual lane are provided, and the method for generating the virtual lane according to an embodiment of the present disclosure comprises: determining a lane recognition limit situation in which a lane in front of an ego vehicle is not recognized; determining whether conditions for entering a virtual lane generation mode are satisfied in the lane recognition limit situation; if the conditions for entering the virtual lane generation mode are satisfied, entering the virtual lane generation mode; processing previous lane information, information of the ego vehicle, and information of a front vehicle; generating the virtual lane based on the processed information; and controlling the ego vehicle based on the generated virtual lane.