NRI: FND: Natural Power Transmission through Unconstrained Fluids for Robotic Manipulation

点击翻译按钮获取中文摘要

This National Robotics Initiative project will promote the progress of science and advance the national health, prosperity and welfare; by studying a novel approach to control the transmission of power from a robot through an unconstrained flow of water to manipulate an object using fluid-structure interaction control strategies and without any other direct contact. In many existing robotic control problems, power is transferred through direct contact with the object to be manipulated. This project focuses on the case of underwater manipulation, where the environment (water) has a significant impact on the overall system performance. In a fluid-based system, interactions between bodies and fluid around them transfer a significant amount of momentum to the water. In this work, an object in water will be controlled by manipulating an upstream structure whose wake can be controlled. The wake of the upstream object will be controlled by forcing it to rotate at a given frequency and angular velocity. Through closed-loop feedback, the desired motion of the downstream object can be obtained by inputting the desired motion of the upstream structure and thus controlling its wake. The approach can be applied to establish a novel method for natural and unconstrained gait training for persons recovering from stroke or injury by naturally assisting their gait using flow forces. Beyond health, extensions of this work can be applied in manufacturing to create new methods for fluid-based material handling and production processes. The project will also motivate and broaden the diversity of the next generation of engineers and scientists through training undergraduate and graduate students and designing and implementing an innovative underwater robotics outreach program for K-12 students. The transformative impact of this work on robotics is enabling a new fluid-based, non-contact manipulation strategy. The work creates new insights in the field of fluid-structure interactions by introducing a closed-loop control method to indirectly impose desired motions of an object placed in the wake of another structure. An “explore-and-exploit” methodology is used to effectively search for parameters in the highly nonlinear system to achieve optimal limit cycle oscillations of a hydrofoil in the wake of a controlled rotating cylinder. The parameters can then be used to design a controller to obtain the desired trajectory for a multi-body system by controlling the rotation rate of an upstream cylinder. With this approach, an underwater robot gait training system is created that can manipulate the power of the existing fluid flow to apply controlled forces to the lower limbs of a person to assist them while walking. The system is supported by new predictive simulations of underwater human gait with fluid-structure interaction dynamics to optimize the interconnected human-robot system. The approach can also be generalized for other applications in underwater robotics for non-contact manipulation using flow forces.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.