Modular Fluidic Propulsion Robots
Matthew J. Doyle,Joo V. Amorim Marques,Isaac Vandermeulen,Christopher Parrott,Yue Gu,Xinyu Xu,Andreas Kolling,Roderich Gro,Joao V. Amorim Marques,Roderich Gros
DOI: https://doi.org/10.1109/tro.2020.3031880
IF: 7.8
2021-04-01
IEEE Transactions on Robotics
Abstract:We propose a novel concept for modular robots, termed modular fluidic propulsion (MFP), which promises to combine effective propulsion, a large reconfiguration space, and a scalable design. MFP robots are modular fluid networks. To propel, they route fluid through themselves. In this article, both hydraulic and pneumatic implementations are considered. The robots move towards a goal by way of a decentralized controller that runs independently on each module face, uses two bits of sensory information and requires neither run-time memory, nor communication. We prove that 2-D MFP robots reach the goal when of orthogonally convex shape, or reach a morphology-dependent distance from it when of arbitrary shape. We present a 2-D hydraulic MFP prototype and show, experimentally, that it succeeds in reaching the goal in at least 90% of trials, and that 71% less energy is expended when modules can communicate. Moreover, in simulations with 3-D hydraulic MFP robots, the decentralized controller performs almost as well as a state-of-the-art and centralized controller. Given the simplicity of the hardware requirements, the MFP concept could pave the way for modular robots to be used at sub-centimeter-scale, where effective modular propulsion systems have not been demonstrated.
robotics