Sensor Array Initialization Using Mahalanobis Distance Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for initializing sensor arrays with multiple measurement axes are unreliable due to the presence of disruptors, which can cause measurement discrepancies that are misinterpreted as errors, leading to incorrect correction coefficients and unreliable measurements.
Innovation Solution
A method that detects the presence of disruptors by calculating Mahalanobis distances between initial measurements and reference values, allowing for the differentiation between intrinsic and extrinsic nonuniformities, and inhibits adjustment of correction coefficients if a disruptor is detected, ensuring accurate initialization and reliable measurements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the first measurements are used to adjust the correction coefficients during initialization, then the measurement discrepancies are compensated, but the reliability of the initialization is reduced when a disruptor is present
Solution Approach 1:
The patent performs a preliminary detection of disruptors using Mahalanobis distance calculation before adjusting the correction coefficients. This preliminary action identifies whether the measurement discrepancies are caused by disruptors or inherent sensor errors, and only adjusts the coefficients if no disruptor is detected, thus preventing unreliable initialization while maintaining measurement precision compensation capability
2Manufacturing precision
If correction coefficients are adjusted based on first measurements, then measurement errors are compensated, but incorrect coefficients are set when disruptors are present
Solution Approach 1:
The patent converts the harmful effect of disruptors into a useful detection mechanism. By calculating the Mahalanobis distance of measurements from the expected uniform distribution, the system identifies disruptor presence and prevents incorrect coefficient adjustment. This transforms the disruptor interference from a purely harmful factor into a detectable signal that protects the initialization process
3Reliability
If a disruptor is detected and coefficient adjustment is inhibited, then measurement reliability is maintained, but initialization time is extended
Solution Approach 1:
The patent implements a feedback mechanism where the Mahalanobis distance calculation continuously monitors measurement uniformity and provides feedback on disruptor presence. This feedback allows the system to make real-time decisions about whether to proceed with coefficient adjustment, maintaining reliability while minimizing unnecessary delays by quickly identifying when disruptors are present
Data Source
AI summary
A method for initializing a sensor array comprises executing a statistical test on a set consisting of first measurements measured by the sensors of the array, the statistical test being able to detect in the first measurements at least one aberrant measurement caused by a presence of an object, called a disruptor, this disruptor modifying the measured physical quantity nonuniformly, wherein if the execution of the statistical test detects at least one aberrant measurement, then the method comprises signaling the presence of the disruptor, and if the execution of the statistical test does not detect at least one aberrant measurement, then the method comprises acquiring second measurements of the physical quantity, the second measurements being measured by the sensors.