NRI-Large: Collaborative Research: Soft Compliant Robotic Augmentation for Human-Robot Teams

该提案涉及实现软机器人系统所需的硬件、控制和规划技术，以提供固有的安全性和适应未来的人机系统。该项目的动机是人机交互的广泛方面，包括用于安全顺应操作的软增强工具，用于神经肌肉疾病康复的软外骨骼，或可以符合任意表面的主动夹紧系统。拟议的研究将解决算法和设备级的挑战，出现在软顺应机器人的设计能够姿态不变和形状不变的把握。结合算法解决方案的建模，控制，规划和适应将导致新的软顺应机械手，不需要精确的几何模型的把握。通过设计柔软的顺应性手指和手，新的方法来把握规划和操纵将成为可能。将开发一种新颖的可组合驱动系统和支持规划和控制算法，这些算法具有受自然肌肉启发的功能。负担得起的软操纵器将为老年人或残疾人提供家庭助手，但这些机器人必须能够像人类一样轻松地操纵自然世界。下一代机器人操纵器还将支持工厂自动化的新水平，其中机器人将与人类协同工作，最终目标是降低美国的制造成本。拟议的软设备将提供一种新的方法来辅助和康复使用的顺应性机器人平台。它们的功能兼容特性将使它们能够与人类并肩工作或作为其身体的一部分，以增强和改善人类的生产力和性能。这种新的可穿戴软机器人技术不仅可以帮助工人执行任务，还可以提高许多人的生活质量。此外，PI具有通过提供各级研究培训（从高中教师和学生到本科生和研究生以及博士后）来整合研究和教育的悠久传统。计划开展一系列活动，向本科生、高中社区、妇女和少数民族进行宣传。

This proposal addresses the hardware, control and planning technologies required to achieve soft robotic systems, in order to offer inherent safety and adaptation to the human-machine systems of the future. The project is motivated by a broad range of aspects of human-robot interaction, including soft augmentation tools for safe compliant manipulation, soft exoskeletons for rehabilitation of neuromuscular disorders, or active clamping systems that can conform to arbitrary surfaces. The proposed research will address the algorithmic and device-level challenges that arise in the design of soft compliant robots capable of pose-invariant and shape-invariant grasping. A combination of algorithmic solutions to modeling, control, planning, and adaptation will lead to new soft compliant manipulators that do not need accurate geometric models for grasping. By designing soft compliant fingers and hands, new approaches to grasp planning and manipulation will be enabled. A novel composable actuation system and supporting planning and control algorithms with features inspired by natural muscle will be developed.Broader Impacts: Soft robots are inherently low-cost. Affordable soft manipulators will enable in-home assistants for the elderly or incapacitated, but these robots must be able to manipulate the natural world as easily as people do. The next generation robot manipulators will also support new levels of factory automation, in which robots will work synergistically with humans with the ultimate goal of reducing the cost of manufacturing in the USA. The proposed soft devices will provide a new approach to assistive and rehabilitative usage of compliant robotic platforms. Their functional compliant properties will enable them to work side-by-side with human beings or as part of their bodies, to augment and improve human productivity and performance. This new wearable soft robotic technology will not only help workers perform tasks, but also improve the quality of life for many people. In addition, the PIs have a long tradition of integrating research and education by providing research training at all levels, from high-school teachers and students to undergraduate and graduate students, and postdocs. A range of activities to reach out to undergraduate students, high-school communities, women and minorities is planned.