Physiology‐Informed Layered Sensing for Intelligent Human–Exoskeleton Interaction

ABSTRACT Wearable exoskeletons offer substantial potential to enhance human mobility in everyday environments. However, their translation beyond laboratory environments remains limited by sensing systems that capture movement but not underlying physiology. Here, we introduce a physiology‐layered perception framework for wearable exoskeletons that aligns sensing and inference with the bone–muscle–skin hierarchy to provide task‐specific physiological cues for assistance, effort optimization, and safety co‐monitoring. The framework maps inertial measurement units (IMUs) to skeletal kinematics, surface electromyography (sEMG) to muscular activation, and strain to cutaneous deformation, with real‐time neural decoders tailored to each objective. As a system‐level demonstration of this framework, we implement it in a soft, lightweight smart‐textile platform integrated with a portable ankle exoskeleton (on‐body sensing and electronics weighing < 20 g). Under leave‐one‐subject‐out evaluation on unseen users, the platform achieves accurate ankle joint‐moment estimation (RMSE = 0.13 Nm kg −1 ), real‐time classification of metabolic trends (96.3% accuracy), and sub‐100 ms detection of injury‐related risk (recall = 0.96). This work establishes a physiology‐aligned sensing architecture that reframes exoskeleton perception from motion tracking to real‐time physiological decoding, offering a pathway toward intelligent, adaptive, and personalized wearable robotics.