External Magnetic Interference Classification in Magnetostrictive Position Sensors using Neuro-Symbolic AI with Log-Likelihood Ratios

Abstract

Magnetostrictive Position Sensors (MPS) are used for precise distance and velocity measurements. They utilize magnetostriction to generate structure-borne sound waves and work on the basis of Time-of-Flight (ToF) calculations. However, external electromagnetic interference (EMI) can impact the accuracy of these sensors by interacting with the magnetic fields of magnetostriction. To address this issue, a novel hybrid approach utilizing both neural and symbolic AI has been developed to classify the intensity of EMI. This system is based on the combination of Log-Likelihood Ratios (LLRs). This study’s findings are particularly significant for industrial environments with numerous sources of external electromagnetic interference, where precise measurement is critical.