Side-Looking Video Sensor for Vehicle Pre-Crash Object Classification

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing vehicle safety systems, particularly those using radar sensors, face challenges in accurately detecting and classifying objects from non-parallel directions, leading to potential false alarms and reliability issues, as they require expensive mechanical adaptations and are not effective with optical sensors.

Innovation Solution

A side-looking video sensor system mounted in the rear view mirror, capable of detecting and classifying objects using full frame information, generates a security signal based on object recognition and features, allowing for earlier detection and reduced false alarms by distinguishing between critical and non-critical objects, and integrating with other sensors for enhanced decision-making.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Difficulty of detecting and measuring

If radar sensors are used for detecting and classifying objects, then detection capability is provided, but classification accuracy deteriorates leading to false alarms

Engineering Contradiction:
Improveobject detection capabilityVSAvoidobject classification accuracy
Core Design Contradiction:
Difficulty of detecting and measuringVSMeasurement precision

Solution Approach 1:

The patent combines radar sensor data with video sensor data to achieve both detection and classification functions. The radar provides early detection of objects while the video sensor provides detailed visual information for accurate classification, eliminating false alarms that occur when using radar alone.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The video sensor acts as an intermediary that clarifies ambiguous radar detections. When radar detects an object, the video sensor provides additional visual context to determine whether it is a true threat or a false alarm source, such as distinguishing between actual obstacles and background elements.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Area of stationary object

If mechanical adaptations are made to radar systems for side-looking detection, then detection coverage is improved, but system complexity and cost increase

Engineering Contradiction:
Improvedetection coverage areaVSAvoidsystem complexity
Core Design Contradiction:
Area of stationary objectVSDevice complexity

Solution Approach 1:

The patent replaces mechanical radar adaptations with an optical video sensor system that provides side-looking detection capability without mechanical moving parts. The video sensor captures visual information from side directions, eliminating the need for complex mechanical adaptations while maintaining detection coverage.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The video sensor creates a visual copy or representation of the side environment, providing detection coverage without requiring physical radar antenna rotations or mechanical adaptations. The optical field of view captures the necessary spatial information.

Inventive Principle:
Principle #26Copying

3Measurement precision

If video sensors are used for object recognition, then classification accuracy is improved, but detection time increases due to full frame processing

Engineering Contradiction:
Improveobject classification accuracyVSAvoiddetection time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system extracts only the necessary visual features from video frames for classification purposes, rather than processing entire frames. By taking out and analyzing only relevant object features, the system maintains high classification accuracy while significantly reducing processing time.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The radar sensor performs preliminary detection to identify potential objects of interest before the video sensor begins detailed analysis. This preliminary action allows the video processing to focus only on regions containing detected objects, reducing overall processing time while maintaining accuracy.

Inventive Principle:
Principle #10Preliminary action

4Ease of manufacture

If side-looking sensors are positioned closer to the vehicle for compact installation, then installation simplicity is improved, but pre-crash detection time is reduced

Engineering Contradiction:
Improveinstallation simplicityVSAvoidpre-crash detection time
Core Design Contradiction:
Ease of manufactureVSLoss of time

Solution Approach 1:

The patent positions the sensor in the rear view mirror, utilizing the vertical and lateral space above the vehicle rather than only horizontal space near the vehicle body. This dimensional change allows the sensor to be positioned optimally for both early detection and compact integration into the vehicle structure.

Inventive Principle:
Principle #17Another dimension (Dimensionality change)

Data Source

PatentEP2070774B1Security system and a method to derive a security signal
Publication Date: 2012.11.07 SMR PATENTS S A R L
  • EP2070774B1 patent drawingFigure 1~2
  • EP2070774B1 patent drawingFigure 3~4
  • EP2070774B1 patent drawingFigure 5

AI summary

This invention relates to a security system for vehicles that have a side-looking pre-crash sensor for anticipating impact by a closing object, such as another motor vehicle, from a direction that is non-parallel to the direction of travel of the host vehicle. Such a system employs optical recognition for acquiring closing objects and for easing classification of the objects. The invention relates further to a method to derive a signal used for example to activate a side air back or to be used and further security systems of a vehicle.