Fabrication and Simulation of Artificial Cilia With Unique Integrated Proprioceptive Sensing Mechanisms
Nina Cao, Massachusetts Institute of Technology
Known for their versatility, efficiency, and collective power, cilia are a strong focus in biologically inspired soft robotics research, with artificial cilia being fabricated, studied, and simulated at great length. Cilia play an important role in a variety of applications, including fluid propulsion, filtering, mixing, as well as object transportation, mechanoreception, chemoreception, and more. This research focuses on the fabrication, sensorization, and simulation of intermediate-scale, pneumatically actuated ciliary systems that integrate segmented control and primarily proprioceptive capabilities. We are developing a coupled platform of simulation and hardware for both proprioceptive and exteroceptive ciliary systems to eventually conduct large-scale analyses of the tradeoffs between distributed and centralized control and also how simulations compare to real physical systems. Here we present results on the fabrication and characterization of two primarily proprioceptive sensors and their integration into artificial cilia: a light sensor and a cracked graphite sensor. We created a modular test setup for pressure controlled sensing to map the deformation of the cilium to signal values, and different signal vs deformation relationships are compared across a variety of fabrication methods. We present results on work to characterize the robustness and reliability of both sensing modalities, and also demonstrate the ability of integrating the light and cracked graphite sensors to operate together in a single cilium. We also show preliminary work on simulating the deformation mechanics of highly nonlinear, hyperelastic materials in “ciliary” geometries, with the goal of extending these simulations to large arrays of cilia coupled with fluidic systems. These simulations will eventually predict the behavior of our artificial ciliary systems in complex scenarios where sensing, response, active learning, and collective behavior are important and inform the future fabrication and use of the actual physical system based on real-time data.