Driver Assistance Blind-Spot Detection Using Object Shadows
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing driver assistance systems fail to detect objects that suddenly emerge from blind spots, leading to unnecessary vehicle deceleration and reduced reliability, thus discouraging the use of driver assistance functions.
Innovation Solution
A driver assistance apparatus that detects shadows in blind spots using image data to estimate the position and movement of potential collision objects, determining the likelihood of collision and adjusting vehicle controls to avoid such objects.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the vehicle control apparatus detects objects in blind spot regions and performs deceleration control, then collision avoidance capability is improved, but false deceleration occurs when no object actually rushes out, reducing reliability and driver acceptance
Solution Approach 1:
The patent introduces shadows as an intermediary indicator to detect objects in blind spots. Instead of directly detecting the object itself in the blind spot region, the system detects shadows cast by objects on visible surfaces (road, sidewalk, wall) as intermediaries. This indirect detection method allows the system to infer object presence and movement in blind spots without requiring direct line-of-sight detection, thereby reducing false alarms while maintaining collision avoidance capability
Solution Approach 2:
The patent replaces direct optical detection of objects in blind spots with shadow-based detection using existing cameras. Instead of relying on complex sensor arrays or LiDAR to directly detect objects in blind spot regions, the system uses standard image capture devices to detect shadows and infer object characteristics. This substitution leverages existing hardware capabilities while achieving blind spot detection functionality
2Loss of information
If the vehicle travels through blind spot regions without detection, then false deceleration is reduced, but the vehicle cannot detect objects that rush out from blind spots
Solution Approach 1:
The patent performs preliminary detection of objects in blind spots by detecting their shadows before the objects actually rush out into the vehicle's path. The system continuously monitors shadow positions, movements, and changes to identify objects that may soon enter the vehicle's trajectory. This preliminary action allows the system to prepare appropriate responses (such as alerting the driver or preparing deceleration) before collision becomes imminent, improving both detection accuracy and reliability
3Difficulty of detecting and measuring
If shadow-based detection is used to detect objects in blind spots, then object detection capability is improved, but the system requires additional processing to estimate light source position and calculate object movement
Solution Approach 1:
The patent makes the existing image capture camera perform multiple functions: it not only captures the surrounding view for general obstacle detection but also detects shadows for blind spot object detection. By analyzing shadow characteristics (position, shape, movement) in the captured images, the system extracts multiple types of information (object presence, light source position, object movement) from a single detection system, reducing the need for additional dedicated sensors while improving detection capability
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
Enables effective detection and avoidance of objects in blind spots, enhancing the reliability and usability of driver assistance systems.
Implementation Method 1
detect a shadow of an object present in a blind spot region as viewed from a vehicle, on the basis of image data that captures a surrounding view around the vehicle
Implementation Method 2
estimate a position of a light source that causes the shadow
Data Source
AI summary
A driver assistance apparatus configured to assist in driving to avoid collision with an object is configured to: detect a shadow of an object present in a blind spot region as viewed from a vehicle, on the basis of image data that captures a surrounding view around the vehicle; estimate a position of a light source that causes the shadow; and calculate a position, a direction of movement, and a speed of movement of the object on the basis of the position of the light source and a time change in the shadow, and determines possibility of collision between the vehicle and the object.


