I-GLIDE: Input Groups for Latent Health Indicators in Degradation Estimation

2025-11-26

AI Summary

Existing methods struggle to disentangle aliased degradation mechanisms in multi-sensor systems and fail to quantify uncertainty in health indicators (HIs). To address this, we propose an input-grouping-driven HI construction framework. First, we introduce projection-path reconstruction into HI generation—integrated with Monte Carlo Dropout and probabilistic latent space modeling—to jointly quantify perceptual (data-driven) and cognitive (model-driven) uncertainties. Second, we propose a sensor-subset grouping paradigm that enables mechanism-specific degradation modeling by leveraging domain knowledge. Evaluated on aerospace and manufacturing system datasets, our method significantly improves remaining useful life (RUL) prediction accuracy and cross-domain generalizability. Moreover, it provides interpretable insights into failure evolution pathways through uncertainty-aware HI trajectories and grouped sensor contributions.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Reasoning under Uncertainty: Uncertainty Representations
• Intelligent Robots: Multimodal Perception & Sensor Fusion

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Accurate remaining useful life (RUL) prediction hinges on the quality of health indicators (HIs), yet existing methods often fail to disentangle complex degradation mechanisms in multi-sensor systems or quantify uncertainty in HI reliability. This paper introduces a novel framework for HI construction, advancing three key contributions. First, we adapt Reconstruction along Projected Pathways (RaPP) as a health indicator (HI) for RUL prediction for the first time, showing that it outperforms traditional reconstruction error metrics. Second, we show that augmenting RaPP-derived HIs with aleatoric and epistemic uncertainty quantification (UQ) via Monte Carlo dropout and probabilistic latent spaces- significantly improves RUL-prediction robustness. Third, and most critically, we propose indicator groups, a paradigm that isolates sensor subsets to model system-specific degradations, giving rise to our novel method, I-GLIDE which enables interpretable, mechanism-specific diagnostics. Evaluated on data sourced from aerospace and manufacturing systems, our approach achieves marked improvements in accuracy and generalizability compared to state-of-the-art HI methods while providing actionable insights into system failure pathways. This work bridges the gap between anomaly detection and prognostics, offering a principled framework for uncertainty-aware degradation modeling in complex systems.

Problem

• Improving health indicator quality for remaining useful life prediction
• Disentangling complex degradation mechanisms in multi-sensor systems
• Quantifying uncertainty in health indicator reliability assessment

Innovation

• Adapts RaPP as health indicator for RUL prediction
• Augments HIs with aleatoric and epistemic uncertainty quantification
• Proposes indicator groups isolating sensor subsets for diagnostics