NRI: FND: Multi-Manipulator Extensible Robotic Platforms

长期以来，机器人技术一直强调“危险、肮脏或枯燥”的任务。该项目旨在提高熟练专家的生产效率，而不是将低技能的大规模制造任务自动化。该项目将实现由领域专家与机器人协作执行的定制、专业和利基任务的自动化。实验室科学家依靠对样本的精确和准确的操作来收集数据。该项目将开发协同机器人自动化工具，使科学探索成为可能。这些工具将包括开源运动平台，可以为不同的应用程序定制，新颖的末端执行器，可以移动物体而不接触它们，以及软件编程环境，可以用于编写将使用这些工具进行的工作流。这些工具允许科学家将精确的操作委托给机器人，而不会失去探索和迭代开发实验和工作流程的能力。这项研究将通过提供与许多科学领域相关的低成本和可访问的基础设施来拓宽和多样化自动化技术的获取途径。该研究将为协同工作流程建立工具，最终用户可以为应用程序特定任务定制这些工具。它将为超声波操作提供新颖的末端执行器，包括操作被捕获的粉末和液体等应用。它将为在多个末端执行器之间自动转换的运动平台提供新的设计模式。这些运动平台将是可定制的，应用专家可以在不需要额外的机器人专业知识的情况下在现场生产。它将提供软件工具，用于控制这些末端执行器和从最终用户角度开发的运动平台，提供抽象，允许领域专家定义和执行协作机器人工作流程。这项工作将通过开发非接触式操作、可定制的协作机器人和自动化工作流程的迭代设计，大大推进机器人技术和人机交互的研究工作。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。

For a long time, robotics has emphasized "dangerous, dirty, or dull" tasks. This project aims to make skilled experts more productive rather than automate lower skilled mass-manufacturing tasks. The project will enable automation of custom, specialized, and niche tasks conducted collaboratively by domain experts together with robots. Laboratory scientists rely on precise and accurate manipulation of samples for the gathering of data. This project will develop co-botic automation tools that enable scientific exploration. These tools will include open-source motion platforms which can be customized for different applications, novel end effectors which can move objects without touching them, and software programming environments that can be used for authoring workflows that will be conducted with these tools. These tools allow scientists to delegate the precise manipulation to a robot without loss of the ability to explore and iteratively develop experiments and workflows. This research will broaden and diversify access to automation technology by providing low-cost and accessible infrastructure that is relevant to many scientific domains.The research will establish tools for co-botic workflows which end-users can customize for application specific tasks. It will contribute novel end-effectors for ultrasonic manipulation, including for applications such as manipulating trapped powders and liquids. It will contribute new design patterns for motion platforms which automatically change between multiple end-effectors. These motion platforms will be customizable, with application experts able to produce them in the field without additional expertise in robotics. It will contribute software tools for controlling these end-effectors and motion platforms developed from an end-user perspective, providing abstractions that allow domain experts to define and execute collaborative robotic workflows. This work will substantially advance research efforts in robotics and human-machine interaction by developing contactless manipulation, customizable co-bots, and iterative design of automated workflows.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.