In-Vehicle Abnormal Sound Localization Using Sound-State Mapping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods struggle to accurately specify the location of abnormal sounds in vehicles, especially when they are intermittent and difficult to reproduce outside specific vehicle conditions, leading to challenges in diagnosing the cause.
Innovation Solution
A method utilizing sound and state variables, including frequency components, torque variations, and vehicle conditions, mapped using a neural network to precisely identify the source of abnormal sounds by replicating the conditions under which the sound is detected.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple microphones are arranged in the vehicle to improve sound detection accuracy, then the ability to specify sound occurrence location is improved, but the cost and device complexity increase
Solution Approach 1:
The patent divides the vehicle into multiple regions (front, rear, left, right) and assigns specific microphones to detect sounds in those regions. The sound occurrence location specifying device segments the sound detection task by determining which region the abnormal sound originates from based on microphone signal strengths, rather than requiring precise localization with multiple microphones throughout the entire vehicle.
Solution Approach 2:
The patent introduces a sound occurrence location specifying device as an intermediary that processes microphone signals and determines sound source regions. This intermediary component translates complex multi-microphone signal analysis into simplified regional location information, reducing the overall system complexity while maintaining detection accuracy.
2Measurement precision
If detailed sound analysis is performed to improve abnormal sound detection accuracy, then the detection precision is improved, but the processing time increases
Solution Approach 1:
The patent performs preliminary actions by continuously monitoring sound signals and maintaining ready-to-use mapping data that correlates sound characteristics with drive-system device states. When an abnormal sound is detected, the system can quickly retrieve pre-established relationships between sound variables and device states, avoiding time-consuming real-time analysis while maintaining detection accuracy.
Solution Approach 2:
The patent applies partial action by focusing analysis only on the specific region where the abnormal sound is detected and only on the drive-system device most likely to produce that sound, rather than analyzing all possible sound sources throughout the vehicle. This selective approach reduces processing time while maintaining detection accuracy.
Data Source
Figure 1
Figure 2A~2B
Figure 3~4
AI summary
There is provided a method of specifying a location of occurrence of an abnormal sound, the method including: storing mapping data in a storage device (83), the mapping data prescribing mapping that receives, as inputs, a sound variable that matches a sound detected in a vehicle and a state variable of a drive-system device of the vehicle synchronized with the sound, and that outputs a location as a main cause of the sound; executing a sound signal acquisition process of acquiring a sound signal output from a microphone (65) that detects a sound; a state variable acquisition process of acquiring the state variable of the drive-system device; and a specifying process of specifying the location of occurrence of the sound corresponding to the sound signal using the sound variable and the state variable as inputs to the mapping. There are also provided a non-transitory storage medium and an in-vehicle device.