EFRI C3 SoRo: Soft, Strong, and Safe Configurable Robots for Diverse Manipulation Tasks

这个项目旨在扩展我们对有效的软机器人设计和控制的基本原理的理解。软机器人和类似肌肉的软致动器与敏捷控制策略相结合，将实现目前仅在自然界中发现的新的操作和运动能力，并允许机器人和人类安全协作。今天的工业机械手可以实现快速和精确的组装，但这些机器人在物理上是隔离的，以确保附近任何人的安全。相比之下，软体机器人的身体由本质上柔软和/或可扩展的材料制成，例如硅橡胶或织物，因此与人类和动物互动是安全的。软体机器人具有连续可变形的结构，具有类似肌肉的驱动机制，模仿生物系统的关键特征，与硬体机器人相比，软体机器人具有相对较多的自由度。软体机器人的能力是当今刚体机器人所无法企及的。例如，软体机器人可以以更自然的方式移动，包括复杂的弯曲和扭曲曲率，而不限于现有机器人操纵器的传统刚体运动学。它们的身体可以连续地变形，理论上提供了无限的自由度，并允许它们根据任务调整形状，例如，适应自然地形或形成包围式抓地力。软机器人也被证明具有快速敏捷的机动能力，并且可以改变它们的刚度来达到特定任务或环境的阻抗。目前对软机器人的设备级和算法方面的研究已经产生了一系列新颖的软设备。该项目将推导出一个系统的数学框架来建模和控制软机器人，并将使用所得到的算法来执行具有各种精度和强度要求的操作任务。研究结果将在制造业、仓库和供应链自动化以及烹饪和清洁等日常家庭活动中具有潜在的用途。这些柔软、坚固、安全的机器人将在老年人或残疾人的辅助护理和物理治疗方面有潜在的应用。该项目利用软机器人的独特功能，延续了首席研究员的外展和教育活动的记录，激发了年轻学生对STEM职业的兴趣。在最近的过去，软机器人社区已经探索了许多不同的组件硬件技术，然而，它们实际使用的基本算法障碍仍然具有挑战性。目前，在刚性和柔性机器人的控制策略之间存在人为的划分；刚性机器人使用接触力和接触几何的高带宽控制，而软机器人几乎完全依赖于由材料特性介导的开环相互作用，来控制产生的力和构型。该项目将通过基于任务保真度的表示，通过软界面的近似动态模型，为软机器人开发基于优化的控制，从而弥合这一差距。这类软、强、安全的机器人将通过内部和接触模型以及相关的规划和控制策略共同开发类肌肉驱动来设计、制造和控制。一种创新的新型人造肌肉设计，通过系统的模块化设计，可以定制执行器来完成特定的任务。建模工作将集中在软机器人与环境的接触丰富的行为上，包括精细的触摸和操作，以及高强度的抓握。这种软与强的结合在软机器人社区尚未得到充分解决，并将允许软机器人在前所未有的应用中安全有效地与人交互。该项目由美国国家科学基金会、新兴前沿和多学科活动办公室（EFMA）和美国空军科学研究办公室（AFOSR）联合赞助。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 seeks to extend our understanding of the principles underlying the design and control of effective soft robots. Soft robots and muscle-like soft actuators coupled with agile control strategies will enable new manipulation and locomotion capabilities currently only found in nature, and allow robots and humans to safely collaborate. Today's industrial manipulators enable rapid and precise assembly, but these robots are physically isolated, to ensure the safety of any humans nearby. In contrast, the bodies of soft robots are made of intrinsically soft and/or extensible materials, such as silicone rubbers or fabrics, and are therefore safe for interaction with humans and animals. Soft robots have a continuously deformable structure with muscle-like actuation that emulates key features of biological systems and provides them with a relatively large number of degrees of freedom as compared to their hard-bodied counterparts. Soft robots have capabilities beyond what is possible with today's rigid-bodied robots. For example, soft-bodied robots can move in more natural ways that include complex bending and twisting curvatures that are not restricted to the traditional rigid body kinematics of existing robotic manipulators. Their bodies can deform in a continuous way, providing theoretically infinite degrees of freedom and allowing them to adapt their shape to their task, for example, conforming to natural terrain or forming enveloping grasps. Soft robots have also been shown to be capable of rapid agile maneuvers and can change their stiffness to achieve a task- or environment-specific impedance. Current research on device-level and algorithmic aspects of soft robots has resulted in a range of novel soft devices. This project will derive a systematic mathematical framework to model and control soft robots and will use the resulting algorithms to perform manipulation tasks with a wide variety of delicacy and strength requirements. The results will have potential uses in manufacturing, warehouse and supply chain automation, and everyday home activities such as cooking and cleaning. These soft, strong, and safe robots will have potential application in assisted care for the elderly or disabled, and for physical therapy. This project uses the unique features of soft robots to continue the Principal Investigators' track record of outreach and educational activities that excite young students about STEM careers.In the recent past, the soft robotics community has explored many different component hardware technologies, however fundamental algorithmic obstacles to their practical use remain challenging. Currently there is an artificial divide between control strategies for rigid and soft robots; rigid robots use high-bandwidth control of contact forces and contact geometry, while soft robots rely almost entirely on open-loop interactions, mediated by material properties, to govern the resulting forces and configurations. This project will bridge this gap by developing optimization-based control for soft robots, via approximate dynamic models of the soft interface, based on representations with a fidelity customized to the task. The proposed class of soft, strong, and safe robots will be designed, fabricated, and controlled by co-developing muscle-like actuation along with internal and contact models and associated planning and control strategies. An innovative new artificial muscle design allows customization of actuators to specific tasks, through systematic modular design. The modeling effort will focus on contact-rich behaviors of the soft robot with the environment, both for delicate touch and manipulation, and for high-force power grasps. Such a combination of soft and strong has not been fully addressed in the soft robotics community and will allow soft robots to interact safely and effectively with people in unprecedented applications.This project is jointly sponsored by the National Science Foundation, Office of Emerging Frontiers and Multidisciplinary Activities (EFMA) and the US Air Force Office of Scientific Research (AFOSR).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.