CAREER: Modular Origami in Soft Robotic Systems

该学院早期职业发展计划（Career）项目研究了将特定机械性能设计成软体机器人的新方法。软机器人是具有高度柔顺物理结构的智能机器，即当施加相对较小的力时，软机器人的身体会弯曲。高顺应性使软体机器人比传统的刚性机器人更安全，也使软体机器人比只有少量离散关节的机器人获得各种形状和以更多方式移动。因为容易弯曲的结构不能承载大的载荷，软机器人的一个主要挑战是创造可变刚度的结构，这些结构只能在特定的方向或特定的时间内弯曲。这个项目提出了一种工程可变刚度软机器人的新方法，灵感来自“折纸”艺术和科学。受折纸启发的软体机器人使用相对坚硬的薄片材料，这些材料可以沿着指定的折线弯曲；折叠模式可以用来创建非常精确和复杂的等效机械性能。这种工程柔性与由橡胶和泡沫等材料制成的软机器人结构形成鲜明对比，这些材料本身固有的柔软性限制了其可实现的强度。在这个项目中，折纸启发的折叠模式被用来创建具有指定兼容行为的模块。然后，将许多这样的模块组合起来，就可以制造出更大、更有能力的机器人。模块化折纸机器人潜力的例子包括在受限空间中柔软的触手式操作，全身和被动的保形抓取，以及可控的形状变化。该项目的成果将促进国家的健康和繁荣，使折纸启发的软体机器人能够用于家庭、医院和制造场所，作为护理人员、同事和自我变形的结构。与该项目相结合的教育活动将使用折纸机器人作为教学工具，通过外展活动、高中实习和本科项目为所有年龄段的学生提供教学工具。该项目将建立模块化和可扩展的设计和制造方法，表征和建模机械性能，并测试轻量级折纸启发的模块化连续体软机器人的控制算法，以智能地执行操作、抓取和形状改变任务。一项全面的综合研究和教育计划将涉及(1)模块化折纸设计、制造和建模，以研究折纸启发模块的力学特性与其设计/制造参数之间的映射关系，使用分布式肌腱进行电寻址驱动，分布式三维变形和外力软传感，以及计算效率高的连续体截面变形的clocloid公式，用于外部载荷下的动态建模；(2)赋予软机器人智能的控制方法，使其能够与环境进行安全的力/位置交互，协同结合状态空间的运动规划和控制，以及大规模模块化系统的超冗余，以最小的驱动和传感要求变形成所需的形状。结果将通过验证案例研究进行评估，包括连续手臂操作，全身包裹和利用保形接触进行抓取，以及非常大自由度折纸驱动的形状变化和刚度控制。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 Faculty Early Career Development Program (CAREER) project studies new ways to engineer specified mechanical properties into soft robots. Soft robots are intelligent machines with a highly compliant physical structure -- that is, the body of the soft robot will bend when relatively small forces are applied to it. High compliance makes soft robots much safer around people than traditional rigid robots, and also allows soft robots to attain a variety of shapes and to move in many more ways than robots with only a small number of discrete joints. Because structures that are easy to bend cannot carry large loads, a major challenge in soft robotics is to create variable-stiffness structures that are compliant only in certain directions, or only at specific times. This project presents a new approach to engineering variable-stiffness soft robots, inspired by the art and science of "origami," or paper folding. Origami-inspired soft robots use thin sheets of relatively stiff materials that are made bendable along designated fold lines; the folding pattern can be used to create very precise and complex equivalent mechanical properties. This engineered flexibility contrasts with soft robot structures made of materials such as rubbers and foams, which are limited in their achievable strength by the intrinsic softness of the material itself. In this project, origami-inspired folding patterns are used to create modules with a specified compliant behavior. Larger and more capable robots can then be built up by combining many such modules. Examples of the potential of modular origami-inspired robots include soft tentacle-like manipulation in confined spaces, whole-body and passive conformal grasping, and controllable shape change. The results of the project will advance the national health and prosperity, by enabling origami-inspired soft robots to be used in homes, hospitals, and manufacturing sites, as care-givers, co-workers, and self-morphing structures. Educational activities integrated with this project will use origami-inspired robots as teaching tools for students of all ages through outreach activities, high-school internships, and undergraduate projects. This project will establish modular and scalable design and fabrication methods, characterize and model mechanical properties, and test control algorithms for lightweight origami-inspired modular continuum soft robots to intelligently perform manipulation, grasping, and shape-change tasks. A comprehensive integrated research and education plan will address (1) modular origami design, fabrication, and modeling, to study the mapping between mechanical properties of origami-inspired modules and their design/fabrication parameters, electrically addressable actuation using distributed tendons, distributed soft sensing of 3-D deformation and external forces, and computationally efficient clothoid formulation of continuum section deformations under external loads for dynamic modeling; (2) Control methods that endow soft robots with intelligence, to enable safe force/position interaction with the environment, synergistically combining motion planning and control at the state space, and hyper-redundancy of large scale modular systems to morph into desired shapes with minimal actuation and sensing requirements. Results will be evaluated through validation case studies in continuum arm manipulation, whole-body wrapping and utilizing conformal contact for grasping, and shape change and stiffness control by very large degree-of-freedom origami actuation.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.