On-Vehicle Component Deterioration Diagnosis via Autonomous Retraining

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing methods for diagnosing the deterioration of on-vehicle components are inefficient, especially when vehicles undergo retrofits, as they require individual optimization of estimation algorithms in manufacturing, leading to increased costs and decreased accuracy over time due to changing vehicle characteristics.

Innovation Solution

A vehicle system that includes a storage device for an estimation algorithm, sensors to detect deterioration parameters, and a control device capable of executing autonomous driving performance tests to update the algorithm based on acquired data, ensuring accurate estimation of component deterioration and facilitating timely replacements.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If individual optimization of estimation algorithms is performed for each vehicle in the manufacturing factory, then the initial estimation accuracy is improved, but the vehicle cost increases and the algorithm becomes outdated after retrofits

Engineering Contradiction:

Improveestimation accuracyVSAvoidvehicle cost

Core Design Contradiction:

Measurement precisionVSEase of manufacture

Solution Approach 1:

The system enables the vehicle to automatically update its own estimation algorithm through autonomous performance tests and data-driven retraining, eliminating the need for manual individual optimization during manufacturing. The vehicle self-adapts to retrofits by collecting new data and retraining the algorithm autonomously, resolving the contradiction between initial accuracy and cost.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The estimation algorithm transitions from a static, pre-optimized state to a dynamic, continuously evolving system. The algorithm is periodically retrained using new data collected from the vehicle's operation, allowing it to adapt to changing conditions after retrofits while maintaining accuracy without incurring additional manufacturing costs.

Inventive Principle:

Principle #15Dynamics

2Measurement precision

If individual optimization of estimation algorithms is performed for each vehicle in the manufacturing factory, then the initial estimation accuracy is improved, but the algorithm accuracy decreases after retrofits due to changing vehicle characteristics

Engineering Contradiction:

Improveestimation accuracyVSAvoidalgorithm accuracy over time

Core Design Contradiction:

Measurement precisionVSReliability

Solution Approach 1:

The system implements a feedback loop where the vehicle continuously collects operational data, evaluates performance against the estimation algorithm, and uses this feedback to retrain and update the algorithm. This closed-loop approach ensures the algorithm remains accurate after retrofits by adapting to new vehicle characteristics through data-driven retraining.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system prepares for future retrofits by establishing a framework for continuous algorithm updates before they are needed. Data collection mechanisms and retraining protocols are set up in advance, enabling the algorithm to quickly adapt when retrofits occur, thus maintaining reliability over time without requiring immediate manual intervention.

Inventive Principle:

Principle #10Preliminary action

3Adaptability or versatility

If manual optimization methods are used for estimation algorithms, then customization is possible, but the process is inefficient and time-consuming

Engineering Contradiction:

Improvealgorithm customizationVSAvoidoptimization efficiency

Core Design Contradiction:

Adaptability or versatilityVSProductivity

Solution Approach 1:

The manual, mechanical process of individual algorithm optimization is replaced with an automated computational system. Machine learning algorithms automatically train and update the estimation models using vehicle data, eliminating the need for manual customization while maintaining adaptability. This substitution dramatically improves productivity by processing data and optimizing algorithms computationally rather than manually.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS12252141B2Vehicle and method for diagnosing deterioration of on-vehicle component

Publication Date: 2025.03.18 TOYOTA JIDOSHA KK

AI summary

A vehicle includes a storage device configured to store an estimation algorithm configured to output a degree of deterioration of a component mounted on the vehicle in response to an input of a value of a parameter related to the component, a sensor configured to detect the value of the parameter, and a control device. The control device is configured to execute a performance test by autonomous driving of the vehicle, acquire data indicating performance of the component during the performance test, and update the estimation algorithm by using the data acquired during the performance test.