Large‐Scale Surface Shape Sensing with Learning‐Based Computational Mechanics. (30th August 2021)

• Record Type: Journal Article
• Authors: Wang, Kui; Mak, Chi‐Hin; Ho, Justin D. L.; Liu, Zhiyu; Sze, Kam‐Yim; Wong, Kenneth K. Y.; Althoefer, Kaspar; Liu, Yunhui; Fukuda, Toshio; Kwok, Ka-Wai
• Is Part Of: Advanced intelligent systems. Volume 3:Number 11(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-08-30
• Subjects: computational mechanics -- ensemble learning -- flexible sensors -- robotic proprioception -- surface shape sensing; Artificial intelligence -- Periodicals; Robotics -- Periodicals; Control theory -- Periodicals; 006.3
• DOI: 10.1002/aisy.202100089
• Languages: English
• ISSNs: 2640-4567

Abstract

Abstract : Proprioception, the ability to perceive one's own configuration and movement in space, enables organisms to safely and accurately interact with their environment and each other. The underlying sensory nerves that make this possible are highly dense and use sophisticated communication pathways to propagate signals from nerves in muscle, skin, and joints to the central nervous system wherein the organism can process and react to stimuli. In a step forward to realize robots with such perceptive capability, a flexible sensor framework that incorporates a novel modeling strategy, taking advantage of computational mechanics and machine learning, is proposed. The sensor framework on a large flexible sensor that transforms sparsely distributed strains into continuous surface is implemented. Finite element (FE) analysis is utilized to determine design parameters, while an FE model is built to enrich the morphological data used in the supervised training to achieve continuous surface reconstruction. A mapping between the local strain data and the enriched surface data is subsequently trained using ensemble learning. This hybrid approach enables real time, robust, and high‐order surface reconstruction. The sensing performance is evaluated in terms of accuracy, repeatability, and feasibility with numerous scenarios, which has not been demonstrated on such a large‐scale sensor before. Abstract : To realize accurate and real‐time proprioception, a hybrid surface‐sensing framework applicable to various sensor shapes and sensing element types is proposed. The framework integrates finite element modeling with machine learning methods, requiring only sparsely distributed sensing elements to predict dynamic and complicated morphology changes of the sensor. Such integration provides basis for the development of large‐scale shape sensing.