Vehicle Lane Change Assist Using Multi-Sensor Object Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle lane change assist systems face inaccuracies in determining lane change execution conditions due to low radar sensor accuracy, leading to potential collisions with non-existent cubic objects being recognized as existent, which can result in unsafe lane changes.
Innovation Solution
The system employs multiple sensors to detect and process cubic object information, using pattern comparison and overlapping detection areas to differentiate between existent and non-existent objects, and limits the number of cubic objects subject to the lane change execution condition determination to reduce computing load and ensure accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If radar sensors are used to detect cubic objects around the own vehicle, then the system can acquire information on other vehicles for lane change execution condition determination, but the low accuracy of radar sensors may cause the system to recognize non-existent cubic objects as existent, leading to inappropriate lane change execution condition determination
Solution Approach 1:
The patent combines detection results from multiple radar sensors (front, rear, left, right) to cross-validate the existence of cubic objects. By merging data from multiple sensors with overlapping detection areas, the system can distinguish between real objects and false detections, thereby improving determination accuracy while compensating for individual sensor precision limitations
Solution Approach 2:
The system implements a feedback mechanism where the lane change execution condition determination is continuously refined based on repeated detection cycles. The determination result feeds back into the detection process, allowing the system to adjust and improve its object recognition accuracy over time, ensuring reliable lane change execution decisions
2Reliability
If the system processes cubic object information from multiple radar sensors to accurately determine lane change execution conditions, then the determination accuracy improves, but the computational load increases
Solution Approach 1:
The patent segments the cubic object information processing into distinct stages: individual sensor detection, overlapping area identification, object existence verification, and lane change condition determination. This segmentation allows the system to process information in manageable portions, reducing the overall computational burden while maintaining high determination accuracy through systematic multi-stage analysis
Solution Approach 2:
The system applies partial processing by focusing computational resources only on cubic objects detected in overlapping areas of multiple sensors, rather than processing all detected objects uniformly. This selective approach reduces unnecessary computations while ensuring that potentially hazardous objects receive thorough analysis, optimizing the balance between accuracy and processing power
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach enhances the accuracy of lane change execution condition determination, preventing unsafe lane changes by correctly identifying non-existent objects and reducing the computational burden, thereby ensuring safe vehicle movement.
Implementation Method 1
Each of the radar sensors transmits a radio wave and receives the radio wave reflected by a cubic object existing around the own vehicle
Data Source
AI summary
A vehicle lane change assist apparatus of the invention executes a lane change assist control for moving an own vehicle to a target next lane when an execution of the lane change assist control is requested by a driver of the own vehicle, and a lane change execution condition is satisfied. The lane change execution condition is a condition that the own vehicle does not contact the one or more recognized cubic objects while the own vehicle is caused to move to the target next lane by the lane change assist control. The vehicle lane change assist apparatus determines that the lane change execution condition is not satisfied when one or more non-existent cubic objects are recognized as the one or more recognized cubic objects.