Vehicle Event Assessment Using Wavelet Transform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing automotive telematics systems face challenges in accurately differentiating between 'real' crashes and mundane driving events, such as hitting a pothole, due to similar sensor readings, which affects the determination of impact event types and requires additional hardware for vibration measurements.
Innovation Solution
A system utilizing a template library and pattern matching processor to analyze motion sensor data, applying wavelet transformations to identify features and compare them with templates, allowing for accurate classification of impact and non-impact events without the need for additional sensors, thereby reducing system complexity and cost.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vibration measurements are used in addition to accelerometer measurements to determine impact event types, then measurement precision is improved, but device complexity increases due to additional sensors
Solution Approach 1:
The patent extracts and utilizes specific frequency components (0.5Hz to 50Hz) from the accelerometer signal through wavelet transformation to represent vibration characteristics. This eliminates the need for separate vibration sensors while capturing the essential vibration information needed for impact event classification.
Solution Approach 2:
The patent replaces the mechanical vibration sensor with a signal processing approach using wavelet transformation on accelerometer data. The wavelet coefficients capture vibration characteristics in the frequency domain, substituting a mechanical sensing system with an electronic signal processing system.
2Reliability
If multiple types of sensors are used to accurately differentiate between crash and mundane driving events, then reliability is improved, but manufacturing cost increases
Solution Approach 1:
The accelerometer serves multiple functions: it detects both impact forces and vibration characteristics. Through wavelet transformation, the same sensor data is processed to extract both crash magnitude information and vibration frequency information, making the single sensor multi-functional.
Solution Approach 2:
The patent changes the parameter representation of the accelerometer signal by applying wavelet transformation. This transforms the time-domain acceleration signal into frequency-domain wavelet coefficients, enabling the extraction of vibration characteristics from the same sensor that measures impact force.
3Device complexity
If a magnitude-based threshold comparison is used to determine impact events, then device complexity is reduced, but measurement precision deteriorates due to inability to differentiate between crash and pothole events
Solution Approach 1:
The patent adds a frequency dimension to the analysis by applying wavelet transformation. Instead of only comparing acceleration magnitude in the time domain, the system analyzes wavelet coefficients across different frequency bands (0.5Hz-50Hz), adding spectral information that enables differentiation between impact types.
Solution Approach 2:
The wavelet transformation is applied preliminarily to the accelerometer signal before classification. This preprocessing step extracts vibration characteristics and organizes the data into frequency-based wavelet coefficients, preparing the data for more accurate impact event differentiation.
Data Source
Figure 1
Figure 2~3
Figure 4a~4b
AI summary
The disclosure relates to apparatus (300) and automated methods (100, 200) for generating a library of templates (304) corresponding to different known types of motor vehicle event and discriminating between types of event on a motor vehicle. The apparatus (300) comprises the template library (304) and a pattern matching processor (302).