Intelligent Soft Wearable Robots for Rehabilitation: A Review of Bioelectronic Sensing, Multimodal Sensor Fusion, and AI-Driven Control
Electronics · 17.11.2026 · 15 (22) · 5355
DOI: 10.3390/electronics15225355 ↗Kurz gefasst
Fachartikel in der Zeitschrift Electronics, online erschienen am 17.11.2026. Verfasst von Lasitha Piyathilaka und 5 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Soft wearable robots transmit assistance through compliant actuators, textiles and body-conforming interfaces, offering opportunities for rehabilitation and the support of residual motor function. Their clinical value depends on the integration of deformable mechanics with reliable sensing, timely state estimation and appropriate control. Structured Scopus and Europe PMC searches informed purposive assessment of selected primary reports. This critical narrative review examines bioelectronic sensing, multimodal sensor fusion and artificial intelligence (AI) within the sensing–estimation–control–feedback loop of soft wearable rehabilitation robots. Upper- and lower-limb architectures, actuation and attachment mechanisms are considered alongside neuromuscular, kinematic, mechanical and physiological measurements. Fusion methods and AI models are assessed in relation to intention recognition, continuous movement estimation, physiological assessment, personalisation and real-time implementation. A five-level control framework distinguishes fixed, sensor-triggered, AI-informed, adaptive intelligent and human–robot co-adaptive assistance, while a complementary evidence framework records integration, population, study design, exposure, deployment setting and assisted or unassisted outcome as separate attributes. Direct soft-wearable evidence is distinguished from supporting work in other rehabilitation robots, prostheses and wearable sensing. Reported advances support selected assisted and training outcomes, but reliable repeated-session operation and retained functional benefit remain unevenly demonstrated. Key barriers include garment-dependent measurement changes, biosignal variability, synchronisation, model generalisation, actuator hysteresis, energy demand, usability and clinical validation. The review identifies priorities for uncertainty-aware fusion, bounded adaptation, local inference, representative benchmarks and longitudinal evaluation, linking technical progress to safe, practical and clinically meaningful rehabilitation.
Zitieren
APA
Piyathilaka, L., Jayawardena, A., Moratuwage, D., Wang, Z., Wijayaraja, J. L., & Sul, J. (2026). Intelligent Soft Wearable Robots for Rehabilitation: A Review of Bioelectronic Sensing, Multimodal Sensor Fusion, and AI-Driven Control. Electronics, 15(22), 5355. https://doi.org/10.3390/electronics15225355
Vancouver
Piyathilaka L, Jayawardena A, Moratuwage D, Wang Z, Wijayaraja JL, Sul J. Intelligent Soft Wearable Robots for Rehabilitation: A Review of Bioelectronic Sensing, Multimodal Sensor Fusion, and AI-Driven Control. Electronics. 2026;15(22):5355. doi:10.3390/electronics15225355
BibTeX
@article{Piyathilaka2026,
author = {Piyathilaka, Lasitha and Jayawardena, Amal and Moratuwage, Diluka and Wang, Zhenglin and Wijayaraja, Jaliya Lakshan and Sul, Jung-Hoon},
title = {Intelligent Soft Wearable Robots for Rehabilitation: A Review of Bioelectronic Sensing, Multimodal Sensor Fusion, and AI-Driven Control},
journal = {Electronics},
year = {2026},
volume = {15},
number = {22},
pages = {5355},
doi = {10.3390/electronics15225355}
}
EUVIMED