AI-Driven Fault Detection in Avionics System Integration Test Rigs

Rifat Iqbal

Abstract: Avionics System Integration Test Rigs (SITRs) validate LRUs, data buses, and flight-critical subsystems before deployment. Traditional test procedures rely on manual inspection and built-in tests that often miss intermittent or evolving faults. We propose an AI-based fault detection framework using multisensor data fusion and LSTM autoencoders to identify anomalies in communication buses (ARINC-429, MIL-STD-1553), power rails, and analog channels. The system learns normal operational patterns and raises real-time anomaly alerts. We demonstrate the method on a simulated avionics SITR dataset with injected faults and show detection accuracy >92%, low false-positive rate, and millisecond-level detection latency. The framework enhances SITR reliability and reduces technician workload.

Keywords: avionics testing, fault detection, sensor data fusion, anomaly detection, LSTM autoencoder

How to Cite?: Rifat Iqbal, "AI-Driven Fault Detection in Avionics System Integration Test Rigs", Volume 15 Issue 5, May 2026, International Journal of Science and Research (IJSR), Pages: 652-657, https://www.ijsr.net/getabstract.php?paperid=SR26510070250, DOI: https://dx.doi.org/10.21275/SR26510070250

Download Citation: APA | MLA | BibTeX | EndNote | RefMan