Blind Spot Detection False Alert Reduction

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing blind spot detection systems struggle to differentiate between a stagnating vehicle and a stationary object, leading to unnecessary warnings and reduced driver trust due to the inability to accurately distinguish between these two conditions.

Innovation Solution

A system utilizing multiple sensors, including radar and video cameras, to detect objects in the blind spot and determine their movement relative to the host vehicle, with additional sensors monitoring areas outside the blind spot to confirm whether the object is stationary or moving, thereby triggering a blind-spot-detection signal only when a stagnating vehicle is present.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If a single sensor type is used for blind spot detection, then the system is simple and cost-effective, but it cannot differentiate between stagnating vehicles and stationary objects

Engineering Contradiction:
Improveobject classification accuracyVSAvoidsensor system complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines multiple sensor types (radar and video camera) into a unified blind spot detection system. The radar sensor detects objects in the blind spot while the video camera captures visual information about additional objects in front of or behind the detected object. By merging these different sensor modalities, the system achieves accurate differentiation between stagnating vehicles and stationary objects, resolving the measurement precision issue while accepting increased device complexity.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The video camera acts as an intermediary sensor that provides additional information about the context of objects detected by the radar. By monitoring areas outside the blind spot (in front of or behind the detected object), the camera mediates the classification process, enabling the system to determine whether a detected object is a stagnating vehicle or a stationary object based on the movement status of surrounding objects.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If the system triggers warnings for all detected objects in the blind spot, then it ensures safety by not missing any potential hazards, but it generates false alerts when stationary objects are detected

Engineering Contradiction:
Improvewarning system reliabilityVSAvoidfalse alert frequency
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The system uses feedback from multiple sensors to validate detections before triggering warnings. The radar detects objects in the blind spot, then the video camera provides feedback about additional objects in front of or behind the detected object. Based on this feedback and the movement analysis, the system intelligently decides whether to trigger a warning, eliminating false alerts from stationary objects while maintaining reliable detection of stagnating vehicles.

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The system dynamically adjusts its warning behavior based on the movement characteristics of detected objects and surrounding objects. By continuously monitoring whether additional objects are moving or stationary, the system adapts its classification and warning generation in real-time, ensuring warnings are only triggered for actual hazards (stagnating vehicles) and not for stationary objects.

Inventive Principle:
Principle #15Dynamics

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

The system effectively differentiates between stagnating vehicles and stationary objects, reducing false warnings and enhancing the reliability of blind spot detection, thus improving driver safety and lane-changing maneuvers.

Implementation Method 1

the one or more sensors include at least one of a radar sensor, a video camera, and an ultrasound sensor

Methodology Applied
Scientific EffectRadar: Radar

Implementation Method 2

the one or more sensors include at least one of a radar sensor, a video camera, and an ultrasound sensor

Methodology Applied
Scientific EffectOptical detection: Photography

Implementation Method 3

the one or more sensors include at least one of a radar sensor, a video camera, and an ultrasound sensor

Methodology Applied
Scientific EffectUltrasound: Ultrasound

Data Source

PatentEP2753523B1Driver assistance system for reducing blind-spot-detection false alerts
Publication Date: 2018.11.14 ROBERT BOSCH GMBH
  • EP2753523B1 patent drawingFigure 1
  • EP2753523B1 patent drawingFigure 2
  • EP2753523B1 patent drawingFigure 3

AI summary

The application relates to detecting objects (303) in the blind spot of a host vehicle (301) and to decide whether a blind spot warning should be issued. When an object (303) is detected in the blind spot of the host vehicle (301), the system analyzes other objects (305, 307) directly in front of or behind the detected object in the blind spot. If the other objects (305, 307) are moving, the system concludes that the object (303) in the blind spot is also moving and, therefore, is a stagnating vehicle. If the other objects (305, 307) are not moving, the system concludes that the object in the blind spot is also a stationary object. This allows recognition of a guardrail (309). The system generates a blind-spot-detection signal when it determines that a stagnating vehicle is located in the blind spot. As alternative, the speed of the object (303) in the blind spot may be determined.