Vehicle Operation Data Collection for Low-Frequency Abnormalities
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in efficiently collecting a sufficient number of abnormal operation data for vehicle abnormality detection, particularly for low-frequency abnormalities, due to communication and analysis load limitations.
Innovation Solution
A vehicle operation data collection apparatus and method that includes a data excess and deficiency evaluation unit to identify insufficient data, extracts target vehicles based on maintenance history, and distributes data collection commands to optimize data acquisition for abnormality detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Quantity of substance
If sensor data from all vehicles is collected to the data center, then the quantity of operation data increases, but communication load, analysis load, and accumulation load increase excessively
Solution Approach 1:
The patent extracts and collects only the necessary operation data that contributes to abnormality detection, rather than collecting data from all sensors in all vehicles. This selective extraction approach reduces communication load while maintaining sufficient data quantity for statistical analysis.
Solution Approach 2:
The patent applies different data collection strategies based on local conditions - collecting data selectively from vehicles based on their operation states and data contribution potential, rather than uniformly collecting all data from all vehicles.
2Quantity of substance
If sensor data from all vehicles is collected to the data center, then the quantity of operation data increases, but analysis load and accumulation load increase excessively
Solution Approach 1:
The patent extracts only the essential operation data that contributes to abnormality detection, filtering out redundant data before transmission to the data center. This reduces both the volume of data requiring analysis and the computational complexity of the analysis process.
3Measurement precision
If abnormal operation data with low occurrence frequency is collected, then the accuracy of abnormality detection improves, but it is difficult to collect sufficient abnormal operation data
Solution Approach 1:
The patent performs preliminary evaluation of data excess and deficiency for each abnormality type before collection. By identifying which abnormality types have insufficient data, the system can proactively target collection efforts toward those specific cases, ensuring sufficient abnormal data is gathered for accurate detection.
Solution Approach 2:
The patent implements a feedback mechanism where the data excess and deficiency evaluation unit continuously monitors data sufficiency and adjusts collection targets accordingly. This feedback loop ensures that collection efforts are dynamically directed toward acquiring the specific abnormal operation data needed to improve detection accuracy.
4Measurement precision
If machine learning is used for abnormality detection, then detection accuracy improves, but data insufficiency for low-frequency abnormalities reduces accuracy
Solution Approach 1:
The patent performs preliminary evaluation of data sufficiency using machine learning before final detection. This preliminary assessment identifies which abnormality types have sufficient training data, allowing the system to apply machine learning accurately where data exists while directing targeted collection efforts for cases with insufficient data.
Data Source
Figure 1
Figure 2A~2B
Figure 3
AI summary
A vehicle operation data collection apparatus includes a vehicle operation history DB which accumulates vehicle operation data acquired from a vehicle; a data excess and deficiency evaluation unit which evaluates excess or deficiency of vehicle operation data accumulated in the vehicle operation history DB for each of abnormality types, on the basis of accuracy information of classification obtained when classifying the abnormality types occurring in the vehicle by machine learning, using vehicle operation data accumulated in the vehicle operation history DB; a collection target vehicle extraction unit which extracts a vehicle suitable for acquiring data of an abnormality type evaluated as data deficiency by the data excess and deficiency evaluation unit from a vehicle maintenance history DB as a collection target vehicle; and a collection command distribution unit which distributes a collection command instructing collection of operation data to the extracted collection target vehicle.