Vehicle Controller Using V2V Performance Indices for Adaptive Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing vehicle controllers diagnose engine abnormalities using fixed threshold values that do not account for varying traveling environments, leading to suboptimal results and potential delays in adjusting vehicle performance.
Innovation Solution
A vehicle controller that performs vehicle-to-vehicle communication to derive and compare traveling performance indices, adjusting relationship specifying data using reinforcement learning to improve vehicle performance and detect potential abnormalities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If fixed threshold values are used for abnormality diagnosis, then the diagnosis system is simple to implement, but the diagnosis accuracy deteriorates because it does not account for varying traveling environments
Solution Approach 1:
The patent transforms the static fixed threshold values into dynamic adaptive thresholds by introducing a learning mechanism. The controller continuously learns from actual vehicle operation data and environmental conditions, automatically adjusting the threshold values to match current traveling environments. This dynamic adaptation resolves the contradiction by maintaining diagnostic simplicity while improving accuracy through environment-specific threshold optimization.
Solution Approach 2:
The patent implements a feedback loop where the controller receives actual operation data, compares it with diagnostic thresholds, and uses the results to refine future threshold values. This closed-loop feedback system allows the diagnosis system to continuously improve its accuracy by learning from past diagnostic outcomes and actual vehicle performance, thereby resolving the contradiction between simple implementation and high diagnostic accuracy.
2Productivity
If fixed threshold values are used for abnormality diagnosis, then the system requires minimal data processing, but the traveling environment context is lost leading to suboptimal diagnosis results
Solution Approach 1:
The patent enables the diagnosis system to self-adjust and self-optimize by automatically learning from operational data without requiring external intervention. The controller autonomously refines threshold values based on accumulated experience and environmental patterns, maintaining data processing efficiency while progressively improving diagnosis reliability through self-learning capabilities.
Solution Approach 2:
The patent dynamically changes the diagnostic threshold parameters based on learned environmental patterns and operational characteristics. Instead of using static parameters, the system adapts threshold values to reflect current traveling conditions, thereby maintaining efficient data processing while significantly improving diagnosis reliability through parameter optimization tailored to specific environments.
3Reliability
If reinforcement learning is used to update relationship specifying data, then the vehicle performance is improved, but the computational complexity and data update time increase
Solution Approach 1:
The patent applies partial reinforcement learning by selectively updating only the most critical relationship parameters rather than recalculating all parameters comprehensively. This partial update approach reduces computational overhead and data update time while still achieving significant performance improvements by focusing on the most impactful relationships between operating conditions and optimal control parameters.
Data Source
AI summary
A vehicle controller is used for a first vehicle and includes processing circuitry. The processing circuitry is configured to execute an index deriving process that derives a traveling performance index of the first vehicle, the traveling performance index being an index related to a traveling performance, an index receiving process that receives the traveling performance index of a second vehicle from the second vehicle through vehicle-to-vehicle communication, and a performance determination process that compares the traveling performance index of the second vehicle with the traveling performance index of the first vehicle to determine whether a traveling performance of the first vehicle is lower than a traveling performance of the second vehicle.