Lane Change Assist Offset Correction for Vehicle Recognition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driving assist systems face challenges in accurately recognizing vehicles in the target lane during lane changes, particularly due to the phenomenon of 'own lane switching,' where the system erroneously determines a vehicle in the target lane as being in the original lane, leading to potential collisions.
Innovation Solution
The system incorporates a lane recognition section, a target detection section, and a control execution section that utilize a front camera and millimeter wave radar to accurately determine the own lane width and deviation, and detect targets around the vehicle. During lane changes, it employs offset processes to correct the extraction of target lane vehicles based on changes in own lane deviation, ensuring proper recognition and collision avoidance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system uses standard lane recognition methods to determine own lane deviation, then the system can identify vehicles in the target lane, but during lane changes the system erroneously switches lane recognition causing vehicles to be misidentified as being in the original lane
Solution Approach 1:
The system performs preliminary actions by detecting lane change intent through the winker lever operation before the actual lane change occurs. It proactively switches from standard lane recognition to offset-based recognition in advance, preventing the misidentification problem before it happens during the lane change process
Solution Approach 2:
The system applies preliminary anti-action by introducing an offset value that counteracts the erroneous lane switching effect. When lane change is detected, the system pre-applies the offset to the own lane deviation value, preventing the target lane vehicle from being incorrectly identified as being in the original lane
2Reliability
If the system prohibits lane change assist control when collision risk is detected, then collision safety is improved, but the system overly restricts lane changes due to erroneous vehicle position misidentification during lane switching
Solution Approach 1:
The system performs preliminary action by switching to offset-based vehicle extraction before the lane change completes. This preliminary switch ensures that vehicles are correctly identified throughout the entire lane change process, allowing safe lane changes to proceed without unnecessary prohibition
Solution Approach 2:
The system applies inversion by reversing the normal vehicle extraction logic during lane changes. Instead of extracting vehicles based on standard lane deviation, it extracts them based on offset-corrected deviation, thereby inverting the erroneous recognition and achieving correct identification
3Device complexity
If the system extracts target lane vehicles using standard own lane deviation values, then the extraction process is simple, but vehicles in the target lane are erroneously determined to be in the original lane during lane changes
Solution Approach 1:
The system performs preliminary action by calculating and storing the offset value based on lane width before vehicle extraction occurs. This pre-calculated offset is then applied during vehicle extraction, maintaining a relatively simple process while improving accuracy
Solution Approach 2:
The system applies parameter changes by modifying the own lane deviation parameter during lane changes. It introduces an offset value that changes the deviation parameter from the standard value to an corrected value, thereby improving vehicle identification accuracy without fundamentally changing the extraction algorithm
Data Source
AI summary
When lane change assist control for assisting a lane change from an own lane (original lane) to an adjacent target lane is performed, the determination as to whether or not another vehicle is in the target lane may be made based on the own lane width, the own vehicle's deviation from the lateral center of the own lane (own lane deviation), and the lateral position of the another vehicle. When the vehicle enters the target lane, own lane switching in which the own lane deviation changes by the lane width occurs and makes it impossible to properly perform the above determination. In view of this, after occurrence of own lane switching, the above determination is made based on a value obtained by adding the own lane width to the another vehicle lateral position or a value obtained by subtracting the own lane width from the another vehicle lateral position.