Software-reprogrammable shape-morphing technology based on distributed actuation / sensing

Abstract

Shape-morphing technology, performing active shape transformation using smart materials, holds great promise for future applications involving micro-machines, transformable displays, unmanned exploration, conformable manipulation, and smart healthcare. The state-of-the-art approaches in the relevant field have focused on realizing complicated shape transformations using innovative manufacturers. In order to achieve multiple shape configurations that are prefixed during the fabrication in general, reprogrammability is of considerable interest, which can reprogram the foldable region and directionality through a specific hardware rectification process. Despite their remarkable success, each in-plant reprogramming restricts the adaptive operation of such technology in unstructured environments. Therefore, we are challenged by developing a new method of realizing the dynamically reprogrammable morphological intelligence that can change the shape of the hardware in a software manner. Given the requirement, this dissertation introduces a methodology called software-reprogrammable (in other words, field-programmable) shape-morphing technology. The realization of the method first involves the development of the hardware having high degree of freedom in the shape-programming process. We propose an electrothermal film that embeds a networked form of resistor that performs a dual functionality of heater and thermoreceptor, selectively recruited through dense electrode connections to configure fold through electronic modulation of its electrical power distribution. Under the collaborative research with KAIST Micro and Nano Transducer laboratory, the material design was devised to induce large, fast, and spatially uniform folding deformation, in a bi-directional manner driven by relative heating and cooling from the temperature for structural neutralization. Given the resistive network, the collection of electronic layout, computational algorithms, and closed-loop control schemes presents intuitive means to blend the user intent in situ on the material shape, yielding a servoed, swift, and robust shape-programming process. To be specific, a programmable actuation based on Genetic algorithm is presented to realize the arbitrary electrical power distribution from the user’s decisions. Simultaneously, the temperature state is in-situ measured in accordance with the distribution of resistance changes, exploiting the interconnected regime that associates the numerical reconstruction scheme, the method which we refer to as Resistive Network Imaging (RNI). The integrated operation of the functionalities serves as a perquisite of closed-loop control on temperature without adding further complexity to hardware design, solely driven by electronically programmable logics with the fast functional transition while avoiding their mutual interference (due to their large differences in electrical power level). All the system is intrinsically driven by embeddable electronics to pave the new way for autonomous system engineering, as potentiated with multi-purpose applications in robotic grasping and locomotion. The future improvements of these capabilities are in the mechanically or geometrically scalable material design (including metamaterials), intrinsic shape sensing, digital twinning toward data-efficient computation, and wireless drives, expanding the scope of versatile engineering methodologies through blending intelligence in morphology and computation.