Inertial Sensor Parametric Degradation Detection via FFT
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Inertial measurement units (IMUs) face challenges in accurately detecting performance degradation of inertial sensors due to environmental changes, such as shock events and rapid vibration profile changes, which can lead to increased navigation errors despite the sensors functioning correctly, as existing Built-In Test (BIT) methods only flag catastrophic failures and not performance degradations.
Innovation Solution
A method involving frequency analysis of inertial sensor data to create frequency profiles, comparing these profiles over time, and using spectral analysis techniques like Fast Fourier Transform (FFT) to detect changes, thereby determining a Parametric Confidence Indicator (PCI) that indicates the confidence in sensor data, allowing for improved navigation and guidance by accounting for degraded sensor performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If frequency analysis and spectral analysis methods are implemented to detect sensor performance degradation, then measurement precision and reliability are improved, but device complexity and computational requirements increase
Solution Approach 1:
The patent performs frequency analysis and spectral analysis in advance to establish baseline frequency profiles of the sensor under normal operating conditions. These pre-established profiles are then used for comparison during operation to detect deviations indicating performance degradation, allowing early detection before catastrophic failure occurs.
Solution Approach 2:
The patent applies partial spectral analysis by focusing on specific frequency ranges and characteristics most relevant to sensor performance degradation. Rather than analyzing the entire spectrum in detail, the method identifies and monitors key frequency markers that indicate sensor health, reducing computational complexity while maintaining detection accuracy.
2Reliability
If continuous monitoring of sensor frequency profiles is performed to detect performance degradation, then reliability is improved, but use of energy increases
Solution Approach 1:
The patent implements periodic monitoring of sensor frequency profiles at strategically selected time intervals rather than continuous monitoring. The system compares frequency profiles at these periodic intervals to detect changes indicating performance degradation, thereby maintaining reliability while significantly reducing the energy consumption associated with continuous signal processing.
3Device complexity
If existing Built-In Test methods are used to detect sensor failures, then device complexity is minimized, but measurement precision deteriorates because only catastrophic failures are detected not performance degradations
Solution Approach 1:
The patent enhances existing BIT methods by adding partial spectral analysis capabilities that focus specifically on detecting performance degradations. Rather than implementing a completely new monitoring system, the approach selectively analyzes specific frequency characteristics that indicate sensor health issues, maintaining simplicity while improving detection precision.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This method enables the navigation system to refine its estimates and respond more effectively to changing mechanical environments, enhancing the accuracy and reliability of inertial navigation by identifying and accounting for sensor performance degradation, even in the absence of hard failures.
Implementation Method 1
comparing these profiles over time, and using spectral analysis techniques like Fast Fourier Transform (FFT) to detect changes
Data Source
Figure 1~2
Figure 3
Figure 4(a)~5(c)
AI summary
A method of determining whether parametric performance of an inertial sensor has been degraded comprises: recording first data output from an inertial sensor; then recording second data output from the inertial sensor; comparing the first data output with the second data output; and determining whether the parametric performance of the inertial sensor has been degraded based on the comparison between the first and second data output.