Autonomous Vehicle Passerby Recognition and Targeted Notification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Passersby experience unease when approaching vehicles in automatic driving mode, as they cannot determine if the vehicle recognizes them, leading to unnecessary notifications when all detected passersby are informed, rather than just those with an uneasy feeling.

Innovation Solution

An automatic driving system that includes a vehicle control calculator, a forward traveling environment recognizer, and a passerby recognition calculator to identify passersby facing the vehicle and notify only those with an uneasy feeling, using a recognition-state notifier to display messages on the windshield.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If all detected passersby are notified that the vehicle recognizes them, then passersby with uneasy feeling are relieved, but unnecessary notifications are generated for passersby who do not need them

Engineering Contradiction:

Improvepasserby safety assuranceVSAvoidnotification accuracy

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The notification system applies different treatment to different passersby based on their individual states. Passersby who are facing the vehicle (indicating potential unease) receive notifications, while those not facing the vehicle do not. This localized differentiation resolves the contradiction by making notifications targeted rather than universal.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

Instead of notifying all passersby by default and then filtering, the system inverts the approach by identifying specific passersby who need notification (those facing the vehicle) and notifying only them. This inversion eliminates unnecessary notifications while ensuring those who need reassurance receive it.

Inventive Principle:

Principle #13The other way round (Inversion)

2Adaptability or versatility

If the system notifies all detected passersby, then no passerby is left without notification, but the notification system becomes less precise and more intrusive

Engineering Contradiction:

Improvenotification coverageVSAvoidpasserby state identification accuracy

Core Design Contradiction:

Adaptability or versatilityVSMeasurement precision

Solution Approach 1:

The system determines notification necessity based on local characteristics of each passerby - specifically whether they are facing the vehicle. This local quality assessment enables precise identification of passersby who may have uneasy feelings, achieving both adaptability and precision.

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system replaces a mechanical blanket-notification approach with an intelligent detection system that uses camera imaging and facial recognition technology to automatically identify passersby who need notification, thereby achieving precise adaptability without manual intervention.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Measurement precision

If the system uses facial recognition to identify passersby facing the vehicle, then notifications are highly targeted, but the system complexity increases

Engineering Contradiction:

Improvepasserby orientation detection accuracyVSAvoidrecognition system complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The imaging device serves multiple functions: it captures images for passerby detection, determines passerby orientation by analyzing facial direction, and provides input for notification decisions. This multi-functionality reduces the need for separate specialized components, thereby managing system complexity while maintaining high detection accuracy.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Solution Approach 2:

The system merges the detection and orientation analysis functions into a unified processing flow where the same imaging data is used for both passerby identification and orientation determination. This consolidation reduces system complexity by eliminating redundant components while maintaining precise identification capabilities.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS10933805B2Automatic driving system

Publication Date: 2021.03.02 SUBARU CORP

AI summary

An automatic driving system includes a vehicle control calculator, a f recognizer, a passerby recognition calculator, and a notifier. The vehicle control calculator causes an vehicle to perform automatic traveling along a traveling route. The recognizer acquires information on a traveling environment ahead of the vehicle. The passerby recognition calculator recognizes a passerby ahead of the vehicle based on the information on the traveling environment. The notifier notifies the passerby of a recognition state that the passerby is recognized, when the passerby is recognized by the passerby recognition calculator. The passerby recognition calculator examines whether the recognized passerby is facing a direction of the vehicle based on information on the passerby, acquired by the recognizer, and specifies the passerby as a notification target passerby when recognizing the passerby facing the direction of the vehicle. The notifier notifies the recognition state to the notification target passerby.