EFRI C3 SoRo: Functional-Domain Soft Robots Precisely Controlled by Quantitative Dynamic Models and Data

该项目将创造一种新型的亚毫米级手术软体机器人，能够在人体中进行不受束缚的手术，并通过设计以不同方式响应外部磁场的软功能材料的可编程域实现。这些材料可以用来推动和引导机器人，或者触发永久的形状变化，根据强迫场的频率选择不同的功能。其他材料领域旨在通过电子特性的变化提供分布式传感。在系统层面上，基于模型和数据驱动的方法将被用于根据实时透视图像和机器人传感器的信息来调制外部磁场以控制体内的机器人。机器人的性能将通过实验验证，包括在动物模型中。与传统技术相比，其结果将提高准确性、可操纵性和可导航性，从而为现有手术机器人或机器人导尿管无法到达的复杂和受限环境提供通道。这些新型机器人将为微创手术开辟新的场所，并可能解决医疗保健领域长期存在的挑战和未满足的需求。该项目将由一组具有互补专业知识的研究人员进行，包括软活性材料设计和制造、本构建模和力学、柔性电子和传感器、机器学习和数据处理、医疗设备设计和转化医学。该项目将为来自代表性不足群体的研究生、本科生和高中生提供研究和培训机会，并将为K-12学生提供讲习班和研讨会。软机器人目前面临着一系列关键挑战，包括无系绳驱动、分布式传感、精确控制和小型化。该项目旨在通过一种范式转换的功能域方法来设计、制造和控制一类新的功能域软机器人（FunDo SoRo），以解决这些挑战。FunDo SoRo具有独立的多功能可编程驱动域，分布式传感和数据驱动策略，用于精确的动力学控制，将代表软机器人设计，制造和控制的新范式。实现FunDo SoRo的具体方法是：1)开发新型功能材料和多材料3D打印技术，通过一组响应静态磁场的集成驱动域、动态磁场下可重构的形状记忆域和能够测量应变、接触压力和温度的传感域，实现基于场的远程驱动、形状重构和分布式传感；2)建立理论模型和计算模型，定量预测FunDo SoRo驱动时的动态响应，并通过机器学习辅助数据驱动策略精确控制FunDo SoRo的动态；3)实验验证亚毫米软连续体机器人用于微创手术，以解决医疗保健领域未满足的需求和挑战，如脑动脉瘤或阻塞性肺疾病。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识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 will create a new class of submillimeter surgical soft robots, capable of untethered operation in the human body, and enabled by programmable domains of soft functional materials designed to respond in different ways to external magnetic fields. These materials can be used to propel and steer the robot or to trigger permanent shape change, with different functions selected by the frequency of the forcing field. Other material domains are designed to provide distributed sensing through changes in electronic properties. At the system level, both model-based and data-driven methods will be used to modulate the external magnetic field to control the robot in the body, based on real-time fluoroscopy images and information from the robot's sensors. Robot performance will be experimentally validated, including in animal models. The result will be enhanced accuracy, steerability, and navigability over conventional techniques, thus providing access to complex and constrained environments unreachable by existing surgical robots or robotic catheters. These new robots will open new venues for minimally invasive surgery and potentially address longstanding challenges and unmet needs in healthcare. The project will be carried out by a team of researchers with complementary expertise, including soft active materials design and fabrication, constitutive modeling and mechanics, flexible electronics and sensors, machine learning and data processing, medical device design, and translational medicine. The project will provide research and training opportunities to graduate, undergraduate and high school students from underrepresented groups, and will offer workshops and seminars for K-12 students.Soft robots are currently facing a set of key challenges including untethered actuation, distributed sensing, accurate control, and miniaturization. This project seeks to address the challenges through a paradigm-shifting functional-domain approach for the design, fabrication, and control of a new class of functional-domain soft robots (FunDo SoRo). FunDo SoRo with self-contained multi-functional domains of programmable actuation and distributed sensing and data-driven strategies for accurate dynamics control will represent a new paradigm in the design, manufacture and control of soft robotics. The specific approaches in achieving FunDo SoRo are to 1) develop novel functional materials and multi-material 3D printing techniques to realize field-based remote actuation, shape-reconfiguration, and distributed sensing through a set of integrated actuation domains responsive to static magnetic fields, shape-memory domains reconfigurable under dynamic magnetic fields, and sensing domains capable of measuring strain, contact pressure, and temperature; 2) develop theoretical and computational models to quantitatively predict the dynamic response of FunDo SoRo upon actuation, and data-driven strategies assisted by machine learning to accurately control the dynamics of FunDo SoRo; and 3) experimentally validate submillimeter soft continuum robots for minimally invasive procedures to address unmet needs and challenges in healthcare such as cerebral aneurysms or obstructive pulmonary diseases.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.