Driver-Adaptive Fault Diagnosis Guidance for Automotive Running Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fault diagnosis systems struggle to guide drivers to a driving pattern that facilitates data acquisition for accurate diagnosis without imposing excessive stress, especially when their normal driving patterns deviate from the optimal pattern.
Innovation Solution
A fault diagnosis support device that generates a recommendation model by setting a boundary on the diagnostic model's available range, creating a driving pattern closer to the driver's normal pattern within the diagnosable range, reducing skill and stress.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the optimal driving pattern for diagnosis is presented to the driver, then the running data suitable for fault diagnosis can be acquired, but the driver experiences excessive skill requirement and stress when their normal driving pattern is far from the optimal pattern
Solution Approach 1:
The system transforms the optimal driving pattern parameters (feature values) into a new recommendation model by adjusting the boundary conditions. When the driver's representative point falls outside the available diagnostic range, the system modifies the recommendation to set the boundary as the new target point, thereby changing the parameters to be within achievable limits while preserving diagnostic value.
Solution Approach 2:
Instead of requiring the driver to reach the optimal driving pattern, the system inverts the approach by generating a new recommendation model that adapts to the driver's actual capabilities. The recommendation point is determined based on the driver's representative point and the boundary of the available range, creating a personalized driving pattern that works with rather than against the driver's normal behavior.
2Measurement precision
If the optimal driving pattern for diagnosis is presented to the driver, then the running data suitable for fault diagnosis can be acquired, but the driver experiences excessive stress when their normal driving pattern is far from the optimal pattern
Solution Approach 1:
The system adjusts the recommendation parameters dynamically based on the driver's representative point position relative to the available diagnostic range. By setting the recommendation point at the boundary when the driver's pattern is outside the range, the system changes the target parameters to be achievable, thereby reducing stress while maintaining diagnostic data quality.
Solution Approach 2:
The system inverts the traditional approach by not forcing the driver to match a fixed optimal pattern, but rather generating a personalized recommendation that accommodates the driver's natural driving style. This inversion reduces psychological stress by making the target feel achievable and personalized rather than imposed.
3Measurement precision
If a fixed optimal driving pattern is used for all drivers, then the fault diagnosis can be performed, but it cannot accommodate drivers whose normal driving pattern is far from the optimal pattern
Solution Approach 1:
The system transitions from a static optimal driving pattern to a dynamic recommendation model that adapts to each driver's characteristics. The recommendation point is calculated based on the driver's representative point and the boundary of the available range, making the system flexible and adaptive to different driving styles while maintaining diagnostic effectiveness.
Solution Approach 2:
The system applies local customization by generating individualized recommendation models for each driver based on their specific representative point. Instead of a universal optimal pattern, each driver receives a tailored recommendation that considers their unique driving characteristics and the boundary conditions of the diagnostic range.
4Measurement precision
If the representative point is outside the available range for diagnosis, then the driver's normal driving pattern cannot be used, but a new recommendation model must be generated
Solution Approach 1:
The system segments the feature value space into the available diagnostic range and identifies the driver's representative point position. By dividing the problem into determining whether the representative point is inside or outside the range, and applying different logic accordingly, the system manages complexity through structured segmentation of the decision process.
Solution Approach 2:
Instead of rejecting drivers whose representative points fall outside the available range, the system inverts the approach by generating a new recommendation model that incorporates the boundary as the target. This inversion transforms an exclusion criterion into an adaptive feature, maintaining diagnostic quality while accommodating diverse driving patterns.
Data Source
Figure 1~2
Figure 3~4
Figure 5~6
AI summary
The purpose of the present invention is to make it easier to guide a driver to drive in an ideal driving pattern in order to eliminate dependency on the driver's driving pattern in a fault diagnosis of an automobile part based on automobile running data, even if the driver's driving pattern is far from the ideal driving pattern. To achieve this purpose, provided is a fault diagnosis support device equipped with: a diagnosis model selection means for outputting a diagnosis model in which, for a feature value used for an examination of an automobile part, an available range available for making a diagnosis and a reference point are stipulated; a driver model generation means for generating, as a driver model, a representative point of the feature value that corresponds to a driver's driving pattern; and a recommendation model generation means for generating, if the representative point is outside the available range, a recommendation model in which a boundary of the available range on the representative point side is set as a recommendation point.