Modeling, Control, and Motion Planning of Magnetic-screw Microrobots in Soft Tissue

本项目通过研究磁控微型机器人，推动科学进步，促进国民健康、繁荣和福利。这些毫米级的微型机器人能够穿过脆弱的人体组织，并可能彻底改变许多医疗程序。通过绕过解剖学障碍，微型机器人有可能提供有针对性的医疗治疗，如药物，诊断材料和/或热消融到目前难以或不可能安全到达的身体部位。这些微型机器人通过使用外部旋转磁场在身体中移动，例如由身体外部的机器人手臂保持和旋转的永磁体。然而，这些微型机器人作为外部磁场运动的函数精确地移动通过组织的方式目前还不足以用于任何临床应用。研究团队将进行基础研究，以提供这些微型机器人如何通过软组织移动所需的知识，如何精确地控制它们沿着所需的路径，以及如何规划安全准确地到达体内所需目标的路径。该研究涉及计算机科学、机械工程、数学建模、控制理论和机器人运动规划等多个学科。研究方法和相关的推广和教育活动将有助于扩大在机器人，计算机科学和机械工程的代表性不足的群体的参与，以及积极影响计算机科学和工程教育。提高对磁螺旋微型机器人力学的理解是该项目第一个研究重点的重点。以前的工作已经大大简化了这些设备的力学，导致高建模不确定性以及由于缺乏完整的理解而被认为不稳定的操作制度。相反，该项目将追求机械建模的范式转变，收集对这些机器人转向行为的完全准确和原则性的理解。具体来说，研究小组将研究一系列越来越复杂的建模方法，根据磁相互作用原理拟合实验收集的数据。对于外部磁场，该项目将考虑均匀场和偶极场。除了建模之外，如何精确地控制和规划这些微型机器人的运动还不清楚;这是第二个研究重点。该项目将首先研究迭代非线性控制方法，为这些机器人创建路径跟踪算法，这些算法对驱动噪声和模型不准确具有自适应性和鲁棒性。此外，该项目将调查这些机器人的运动规划方法，这些方法将考虑到它们的能力，利用学习状态采样，并解释它们的力学不确定性。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

This project promotes the progress of science and advances the national health, prosperity, and welfare by investigating magnetically-controlled microrobots. These millimeter-scale microrobots are able to steer through delicate human tissue and may revolutionize many medical procedures. By steering around anatomical obstacles, the microrobots have the potential to deliver targeted medical treatment, such as drugs, diagnostic materials, and/or thermal ablation to sites in the body that are currently difficult or impossible to reach safely. These microrobots move through the body via the use of an external rotating magnetic field, such as a permanent magnet held and rotated by a robot arm on the outside of the body. However, the way in which these microrobots precisely move through tissue as a function of the external magnetic field’s motion is not currently sufficiently understood for use in any clinical application. The research team will pursue fundamental research to provide needed knowledge in how these microrobots move through soft tissue, how to precisely control them along desired paths, and how to plan paths that safely and accurately reach desired targets in the body. This research involves several disciplines including computer science, mechanical engineering, mathematical modeling, control theory, and robot motion planning. The research approach and associated outreach and educational activities will help to broaden participation of underrepresented groups in robotics, computer science, and mechanical engineering, as well as positively impact computer-science and engineering education.The technical aims of the project are divided into two thrusts. Improving understanding of magneticscrew microrobots’ mechanics is the focus of the first research thrust of this project. Prior work has significantly simplified the mechanics of these devices, resulting in high modeling uncertainty as well as operating regimes that were deemed unstable due to lack of a complete understanding. The project will instead pursue a paradigm shift in mechanical modeling, gathering a fully accurate and principled understanding of the steering behavior of these robots. Specifically, the research team will investigate a sequence of increasingly complex modeling methods, fitting to experimentally gathered data, based on magnetic interaction principles. For the external magnetic field, the project will consider both uniform fields and dipole fields. Beyond modeling, how to accurately control and plan motions for these microrobots is not yet understood; this is the focus of the second research thrust. The project will first investigate iterative nonlinear control methods to create path-following algorithms for these robots that are adaptive and robust to actuation noise and model inaccuracies. Additionally, the project will investigate motion-planning methods for these robots that account for their capabilities, leverage learned state sampling, and reason about uncertainty in their mechanics.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.