Blind Spot Monitoring With Road Edge-Based Alert Suppression
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Blind spot monitoring systems often generate false positive alerts due to radar sensors interpreting static objects like guardrails or parked vehicles as moving objects, leading to unnecessary warnings for drivers.
Innovation Solution
A vehicle sensing system that combines camera and radar sensor data using an electronic control unit (ECU) to process image and radar data, determining the vehicle's position relative to road edges and suppressing false positive alerts by using polynomial coefficients and confidence values from image data to validate the presence of static objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If radar sensors are used to detect objects in blind spots, then detection capability is improved, but false positive alerts increase due to misinterpretation of static objects
Solution Approach 1:
The patent introduces camera-based vision data as an intermediary to validate radar detections. The system uses polynomial coefficients from camera image processing to represent road geometry, which then serves as a reference to distinguish between actual moving objects and static false targets like guardrails. This intermediary validation layer resolves the contradiction by filtering radar alerts through additional contextual information.
Solution Approach 2:
The system implements feedback by continuously comparing radar-detected objects against road geometry models derived from camera data. When a radar detection conflicts with the expected road geometry (e.g., an object appears where the polynomial model indicates only road edge should exist), the system uses this feedback to suppress false alerts. This closed-loop validation improves reliability while maintaining detection sensitivity.
2Measurement precision
If polynomial coefficients are used to represent road geometry, then lane and road edge determination accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent transforms complex road geometry representations into simplified polynomial coefficients that capture essential lane and road edge information. By changing the parameter representation from raw pixel coordinates to polynomial form, the system achieves accurate road geometry modeling with reduced computational burden during validation operations.
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
Effectively reduces false positive blind spot warnings by accurately differentiating between static road features and moving objects, enhancing driver safety during lane changes and merging scenarios.
Implementation Method 1
The camera includes a CMOS imaging array that includes at least one million photosensors arranged in rows and columns
Implementation Method 2
The radar sensor is disposed at the equipped vehicle and sensing at least rearward and sideward of the equipped vehicle. The radar sensor is operable to capture radar data
Data Source
AI summary
A vehicular sensing system includes a camera and a radar sensor disposed at a vehicle. The vehicular sensing system, responsive to processing of radar data, determines presence of an object within a blind spot of a driver of the equipped vehicle that is at least sideward of the equipped vehicle. The vehicular sensing system, responsive to processing of image data, determines that the equipped vehicle is within a traffic lane that borders an edge of a road. The vehicular sensing system, responsive to determining that the detected object is not sideward of a side of the equipped vehicle that is closest to the edge of the road, generates a blind spot warning. The vehicular sensing system, responsive to determining that the detected object is sideward from the side of the equipped vehicle that is closest to the edge of the road, suppresses the blind spot warning.