{"slug": "digital-twin-driven-adaptive-sim-to-real-alignment-via-reinforcement-learning", "title": "Digital Twin-Driven Adaptive Sim-to-Real Alignment via Reinforcement Learning for Vibration-Based Bearing Health Monitoring Under Data Scarcity", "summary": "Researchers developed a reinforcement learning-based method for digital twin-driven bearing health monitoring that adaptively aligns simulated and real vibration data, achieving 92.8% cross-equipment diagnostic accuracy without retraining. The approach formulates feature alignment as a Markov decision process solved via Proximal Policy Optimization, addressing heterogeneous sim-to-real gaps across fault classes under data scarcity. Validation on three testbeds demonstrates superior transferability over existing domain adaptation techniques.", "body_md": "arXiv:2606.24954v1 Announce Type: new\nAbstract: Vibration-based health monitoring of rotating machinery requires reliable fault diagnosis under operational data constraints, yet condition assessment remains challenged by structural scarcity of fault events and heterogeneous sim-to-real gaps in digital twin-generated signals. Each fault type generates impulses with distinct periodicity, amplitude modulation, and spectral character, making feature-space discrepancies fundamentally heterogeneous across fault classes. Existing domain adaptation methods apply a class-agnostic global transformation that cannot close all fault-specific gaps without distorting inter-class separability, while uniform source-target mixing introduces distributional noise into the data-abundant Normal class. These limitations stem from treating a sequential, state-dependent alignment problem as a one-shot optimization. Each corrective transformation simultaneously reshapes all class distributions, creating state dependencies that static gradient descent cannot resolve. We formulate feature alignment as a continuous-action Markov decision process solved via Proximal Policy Optimization, where the learned policy issues fault-type-specific affine corrections responsive to the current feature-space configuration, with a dual-objective reward balancing gap minimization against separability preservation. An asymmetry-aware strategy reserves real data for the Normal class while augmenting fault classes with policy-aligned simulated samples. Validation across XJTU-SY, CWRU, and a self-built slewing bearing testbed confirms the dominant gain from reinforcement learning-driven alignment, and cross-equipment linear probing achieves 92.8% without encoder retraining, demonstrating transferable monitoring capability.", "url": "https://wpnews.pro/news/digital-twin-driven-adaptive-sim-to-real-alignment-via-reinforcement-learning", "canonical_source": "https://arxiv.org/abs/2606.24954", "published_at": "2026-06-25 04:00:00+00:00", "updated_at": "2026-06-25 04:26:01.245931+00:00", "lang": "en", "topics": ["machine-learning", "ai-research"], "entities": ["XJTU-SY", "CWRU", "Proximal Policy Optimization", "Markov decision process"], "alternates": {"html": "https://wpnews.pro/news/digital-twin-driven-adaptive-sim-to-real-alignment-via-reinforcement-learning", "markdown": "https://wpnews.pro/news/digital-twin-driven-adaptive-sim-to-real-alignment-via-reinforcement-learning.md", "text": "https://wpnews.pro/news/digital-twin-driven-adaptive-sim-to-real-alignment-via-reinforcement-learning.txt", "jsonld": "https://wpnews.pro/news/digital-twin-driven-adaptive-sim-to-real-alignment-via-reinforcement-learning.jsonld"}}