AI Judder Diagnosis in Mobile GDS for Geometric vs Friction Faults
Solution Overview
Problem
Conventional tablet PC-based diagnostic equipment for vehicle judder diagnosis relies on indirect methods using vibration sensor values from randomly positioned sensors, leading to unreliable results due to sensor deviation and limited ability to distinguish between geometric and friction judder phenomena.
Innovation Solution
A vehicle judder diagnostic method using artificial intelligence and a mobile-based Global Diagnostic System (GDS) that samples input data from multiple sensors mounted in the vehicle, builds a deep neural network (DNN) model, and diagnoses judder phenomena by classifying them as geometric or friction judder based on sensor signal analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If indirect diagnostic method using vibration sensor values from randomly positioned sensors is used, then diagnostic equipment complexity is reduced, but measurement precision and reliability deteriorate due to sensor deviation and disturbance
Solution Approach 1:
The patent segments the diagnostic approach by dividing sensor placement into two categories: random positioning for portable equipment and predetermined optimal positions for mounted sensors. This segmentation allows the system to leverage mounted sensors' precise positioning data while maintaining the option for portable diagnostic tools, thereby resolving the contradiction between equipment simplicity and measurement accuracy.
Solution Approach 2:
The patent introduces an intermediary processing layer that receives data from both randomly positioned vibration sensors and mounted sensors. This intermediary layer applies signal processing and fusion algorithms to compensate for the limitations of random sensor positioning, effectively bridging the gap between simple portable equipment and precise mounted sensor systems.
2Ease of operation
If conventional judder diagnostic software is used, then ease of operation is maintained, but the ability to distinguish between geometric and friction judder deteriorates
Solution Approach 1:
The patent changes the analytical parameters by introducing multiple sensor data streams with different characteristics (mounted vs. portable sensors, vibration acceleration, frequency spectra). These parameter changes enable the diagnostic system to differentiate between geometric and friction judder causes while maintaining user-friendly operation through automated analysis algorithms.
Solution Approach 2:
The patent adds another dimension to the diagnostic analysis by incorporating spatial information from mounted sensor positions and temporal frequency characteristics. This multi-dimensional approach allows the system to classify judder causes more accurately without complicating the user interface, as the complex analysis is handled automatically by the software.
3Device complexity
If acceleration sensor signal values with low sampling rate and resolution are used, then device complexity is reduced, but measurement precision deteriorates
Solution Approach 1:
The patent merges data from multiple sensor sources including mounted acceleration sensors with lower specifications and portable vibration sensors. By combining these data streams and applying signal fusion techniques, the system compensates for the low sampling rate and resolution of individual sensors, achieving high measurement precision without requiring complex high-specification sensor systems.
Data Source
AI summary
A vehicle judder diagnostic method using artificial intelligence applied to a mobile-based GDS according to the present disclosure is characterized in that the mobile-based GDS samples a plurality of sensor signals of a sensor mounted in a vehicle in a vehicle during operation in a judder evaluation mode to quickly, separately diagnose whether the judder phenomenon of the vehicle is a geometric judder or a friction judder by mounting a deep neural network (DNN) model, developed by the trial and error process of a DNN by using the plurality of sensor signals of a test vehicle mounted with a double clutch transmission (DCT), as a judder determination artificial intelligence model 30 in the mobile-based GDS.