Forward Collision Warning False Alert Suppression

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Forward Collision Warning (FCW) systems often generate false alerts due to radar signals reflecting from traffic signs or metallic objects, leading to unnecessary warnings and potential safety risks, as well as driver annoyance and loss of trust in the system.

Innovation Solution

A method and system that determine the presence of a target object within a host vehicle's path, check if the object is stored in a false alert database, and suppress warnings for objects classified as ground or overhead structures, with the option to delay warnings and update the database based on driver reactions, using a combination of sensors and a remote database for accurate classification and learning.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If radar signals are used to detect target objects, then collision detection capability is improved, but false positive alerts increase due to reflections from traffic signs or metallic objects

Engineering Contradiction:
Improvecollision detection capabilityVSAvoidfalse positive alerts
Core Design Contradiction:
ReliabilityVSObject-generated harmful factors

Solution Approach 1:

The system performs preliminary classification of detected objects using multiple sensors (camera, radar, LIDAR) before generating collision warnings. By pre-classifying objects as vehicles, pedestrians, cyclists, or stationary objects, the system identifies false targets before they trigger warnings, thereby reducing false positives while maintaining detection capability

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system introduces an intermediary classification layer between radar detection and collision warning generation. This intermediary process uses multi-sensor data fusion and object classification algorithms to filter out false targets (traffic signs, metallic objects) before they can trigger false warnings, resolving the contradiction between detection sensitivity and false alarm reduction

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If FCW system generates warnings for all detected target objects, then safety coverage is improved, but driver trust decreases due to frequent false alerts

Engineering Contradiction:
Improvesafety coverageVSAvoiddriver trust
Core Design Contradiction:
ReliabilityVSEase of operation

Solution Approach 1:

The system applies different quality thresholds and warning criteria for different types of detected objects. High-priority warnings are generated only for vulnerable road users (pedestrians, cyclists) and moving vehicles, while stationary objects require multiple detections or higher confidence thresholds. This localized quality approach maintains comprehensive safety coverage while reducing false alerts that undermine driver trust

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The system incorporates driver feedback mechanisms where driver responses to warnings (acknowledgment, correction, or no response) are used to adjust future warning behavior. When drivers consistently ignore or correct warnings, the system learns to suppress similar false warnings, thereby maintaining safety coverage while adapting to driver preferences and reducing trust-eroding false alerts

Inventive Principle:
Principle #23Feedback

3Reliability

If autonomous emergency braking is triggered by FCW system, then collision prevention is improved, but safety risk increases when false braking occurs due to tailgating vehicles

Engineering Contradiction:
Improvecollision preventionVSAvoidsafety risk from false braking
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The system performs preliminary verification of target object validity before triggering autonomous emergency braking. Multiple classification checks, sensor cross-validation, and false target identification algorithms are executed in advance to ensure the detected object is a genuine collision threat and not a false target, thereby preventing false braking while maintaining collision prevention capability

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The system implements a multi-stage warning and verification process before initiating autonomous emergency braking. Graduated warning levels and confirmation checks serve as a cushioning mechanism to prevent premature or false braking actions, allowing the system to distinguish between genuine threats requiring immediate braking and false targets that would cause dangerous false braking

Inventive Principle:
Principle #11Beforehand cushioning (Prior cushioning)

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 significantly reduces false positive alerts, enhancing the reliability and safety of FCW systems by distinguishing between actual and false threats, thereby maintaining driver trust and preventing unnecessary braking or system shutdown.

Implementation Method 1

FCW systems use information from the outside world through radar, laser based sensors, or cameras to detect if a target object is obstructing the path of the host vehicle

Methodology Applied
Scientific EffectRadar: Radar

Implementation Method 2

the radar signals may reflect from traffic signs or large metallic objects in the roadway on the ground

Methodology Applied
Scientific EffectReflection: Reflection

Implementation Method 3

FCW systems use information from the outside world through radar, laser based sensors, or cameras to detect if a target object is obstructing the path of the host vehicle

Methodology Applied
Scientific EffectLIDAR: LIDAR

Data Source

PatentEP3134888B1False warning reduction using location data
Publication Date: 2018.07.18 ROBERT BOSCH GMBH
  • EP3134888B1 patent drawingFigure 1
  • EP3134888B1 patent drawingFigure 2
  • EP3134888B1 patent drawingFigure 3

AI summary

A collision warning system for a host vehicle includes a false warning reduction arrangement. A collision sensor senses a presence of a target object within a forward path of a vehicle and a navigation system senses the global position of the vehicle. The collision warning system receives the location information and the target object to provide a collision warning to a vehicle operator. To avoid false alerts, the collision warning system compares target object and location to target objects and locations stored in a false alert database. When the target object is provided in the database, a warning is delayed or suppressed. The collision warning system also senses driver reactions to a warning. When a warning results in no reaction by a vehicle operator, the warning is considered a false alert, and the target object and location are stored in the false alert database.