Vehicle Device Diagnosis Using Nearby-Vehicle Learning Data

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing diagnosis technologies for vehicle-mounted devices struggle to accurately diagnose abnormalities due to environmental changes such as road surface shape, gradient, and weather, which affect the reliability of diagnostic data.

Innovation Solution

A diagnosis device that acquires learning information from nearby vehicles within a predetermined range, calculates a learning value for a target vehicle using reliable data from multiple sources, and compares detection values with calculated learning values to determine operational states, thereby accurately diagnosing abnormalities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If diagnosis is performed using data from a single device, then device complexity is reduced, but measurement precision deteriorates due to environmental variations

Engineering Contradiction:
Improvediagnostic accuracyVSAvoiddata processing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent combines measurement data from multiple devices of the same type mounted in different vehicles to create a cluster set. This merging of data sources improves diagnostic accuracy by compensating for environmental variations that affect individual devices, while the automated clustering algorithm manages the complexity of processing multiple data sources.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent creates virtual copies of device behavior patterns through clustering, where each cluster represents a typical operational pattern under specific environmental conditions. These cluster-based models serve as reference copies for comparing against actual device measurements, improving diagnostic precision without requiring direct physical replication of devices.

Inventive Principle:
Principle #26Copying

2Measurement precision

If data from multiple vehicles is collected to improve diagnosis accuracy, then measurement precision improves, but loss of time increases due to data acquisition and processing

Engineering Contradiction:
Improvediagnostic accuracyVSAvoiddata processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary clustering of historical measurement data from multiple devices to establish reference cluster sets before actual diagnosis is needed. This advance preparation of data patterns enables faster real-time diagnosis by comparing current measurements against pre-organized clusters, reducing the time penalty of processing multiple data sources.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent replaces manual or rule-based data processing with automated machine learning clustering algorithms that can efficiently process large volumes of multi-vehicle data. This computational approach substitutes traditional time-consuming analytical methods with algorithms that automatically identify patterns and generate diagnostic conclusions from multiple data sources.

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

3Measurement precision

If environmental factors are considered in diagnosis, then measurement precision improves, but device complexity increases due to additional processing requirements

Engineering Contradiction:
Improvediagnostic accuracyVSAvoidprocessing complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies different processing approaches to different aspects of the data: environmental conditions are used to group devices into clusters, while device measurements are compared against cluster-specific patterns. This localized processing strategy improves diagnostic accuracy by tailoring analysis to environmental contexts without requiring complex uniform processing of all data elements.

Inventive Principle:
Principle #3Local quality

Solution Approach 2:

The patent creates a universal clustering framework that can handle various device types and environmental conditions through a single standardized process. The cluster-based approach provides a multi-functional solution that adapts to different scenarios without requiring separate complex processing systems for each device type or environmental condition.

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3547058B1Diagnosis device and diagnosis method
Publication Date: 2022.06.29 TOYOTA JIDOSHA KK
  • EP3547058B1 patent drawingFigure 1
  • EP3547058B1 patent drawingFigure 2
  • EP3547058B1 patent drawingFigure 3

AI summary

A diagnosis device (10) includes a learning information acquisition unit (110) configured to acquire a learning value of a first device (30) mounted in at least one first vehicle (1-2, 1-3) present in a predetermined range from a second vehicle (1-1), a learning unit (120) configured to calculate a learning value of a second device (30) mounted in the second vehicle (1-1) using the learning value of the first device (30), and a diagnosis unit (130) configured to diagnose an operation state of the second device (30) by comparing a detection value of the second device (30) with the learning value of the second device (30).