Collaborative Research: RI: Medium: Robust Assembly of Compliant Modular Robots

合作研究：RI：中：兼容模块化机器人的稳健组装

• 批准号: 1956027
• 负责人: Kostas Bekris
• 金额: $ 32.3万
• 依托单位: Rutgers University New Brunswick
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2020
• 资助国家: 美国
• 起止时间: 2020-10-01 至 2024-09-30
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1956027&HistoricalAwards=false
• 关键词: Collaborative; Research; RI; Medium; Robust

This project explores how flexible robots can be designed to move and join together to form larger structures, such as temporary antennas, tent supports, bridges, or tunnel reinforcements. The importance of the project lies in the development of lightweight robots and large-scale systems of robots that can be deployed quickly to form temporary infrastructure. Novel designs of flexible robots, and the hardware and software systems to control them, advance the fields of robotics, automated control, and mechanical design. Societal benefits include humanitarian aid and disaster relief. Disasters ranging from virus outbreaks to coastal storms require quick deployment of infrastructure, such as wireless antennas for restoring communication, scaffolding for restoring power or tents for impromptu patient care. The broader impact of the project will be supported through the development and sharing of open-source design kits and software packages to help with experiment replication and use in hands-on educational activities, as well as interdisciplinary workshops and outreach activities to involve members of underrepresented groups.The design of these robots centers around the principle of tensegrity: structures that contain rigid components that provide structural integrity, and flexible components that distribute forces and allow robots to adapt in shape to their environment. The primary contribution of the project will be an end-to-end exploration of the mechanical designs, state estimation, planning and control systems that enable a novel class of flexible modular robots that deploy and assemble to form larger structures. The project will design tensegrity robots with predictable and capable locomotion capabilities, as well as with docking mechanisms to allow the formation of larger structures from systems of robots. Simulation-driven design will inform what geometric and physical properties modules and resulting structures should and can achieve. The project will explore state estimation techniques that discover the current geometric configuration of flexible robots in contact with each other and the environment. Integrated planning and adaptive control techniques will build on state estimation to allow fine motion skills required for locomotion and docking. The work will be experimentally evaluated at each stage with physical prototype designs that achieve subgoals ranging from first simple locomotion to assembly of flexible structures.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.

这个项目探索如何设计灵活的机器人来移动和连接在一起，形成更大的结构，如临时天线、帐篷支撑、桥梁或隧道加固。该项目的重要性在于开发可以快速部署的轻型机器人和大型机器人系统，以形成临时基础设施。柔性机器人的新设计，以及控制它们的硬件和软件系统，推动了机器人技术、自动化控制和机械设计领域的发展。社会效益包括人道主义援助和救灾。从病毒爆发到沿海风暴等各种灾害都需要快速部署基础设施，例如用于恢复通信的无线天线、用于恢复电力的脚手架或用于临时护理患者的帐篷。将通过开发和共享开源设计工具包和软件包来支持该项目的更广泛影响，以帮助实验复制和在实践教育活动中使用，以及跨学科研讨会和外联活动，让代表性不足的群体成员参与进来。这些机器人的设计围绕着张拉整体原理：结构包含刚性组件，提供结构完整性，柔性组件，分配力，使机器人能够适应其环境的形状。该项目的主要贡献将是对机械设计，状态估计，规划和控制系统的端到端探索，使新型灵活模块化机器人能够部署和组装成更大的结构。该项目将设计具有可预测和有能力的运动能力的张拉整体机器人，以及对接机制，以允许从机器人系统形成更大的结构。仿真驱动的设计将告知模块和最终结构应该和可以实现的几何和物理属性。该项目将探索状态估计技术，以发现与彼此和环境接触的柔性机器人的当前几何构型。综合规划和自适应控制技术将建立在状态估计的基础上，以允许运动和对接所需的精细运动技能。这项工作将在每个阶段通过物理原型设计进行实验评估，以实现从第一个简单运动到柔性结构组装的子目标。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。

项目成果

6N-DoF Pose Tracking for Tensegrity Robots

张拉整体机器人的 6N-DoF 位姿跟踪

• 发表时间: 2022
• 期刊: The International Symposium of Robotics Research
• 作者: Lu, Shiyang;Johnson III, William R;Wang, Kun;Huang, Xiaonan;Booth, Joran;Kramer-Bottiglio, Rebecca;Bekris, Kostas
• 通讯作者: Bekris, Kostas

StarBlocks: Soft Actuated Self-Connecting Blocks for Building Deformable Lattice Structures

• DOI: 10.1109/lra.2023.3284361
• 发表时间: 2023-08
• 期刊: IEEE Robotics and Automation Letters
• 影响因子: 5.2
• 作者: Luyang Zhao;Yijia Wu;Wenzhong Yan;Weishu Zhan;Xiaonan Huang;Joran W. Booth;Ankur M. Mehta;Kostas E. Bekris;Rebecca Kramer‐Bottiglio;Devin J. Balkcom

Sim2Sim Evaluation of a Novel Data-Efficient Differentiable Physics Engine for Tensegrity Robots

用于张拉整体机器人的新型数据高效可微物理引擎的 Sim2Sim 评估

• 发表时间: 2021
• 期刊: IEEE/RSJ International Conference on Intelligent Robots and Systems
• 作者: Wang, Kun;Aanjaneya, Mridul;Bekris, Kostas E
• 通讯作者: Bekris, Kostas E

A Recurrent Differentiable Engine for Modeling Tensegrity Robots Trainable with Low-Frequency Data

• DOI: 10.48550/arxiv.2203.00041
• 发表时间: 2022-02
• 期刊: 2022 International Conference on Robotics and Automation (ICRA)
• 作者: Kun Wang;Mridul Aanjaneya;Kostas E. Bekris

Real2Sim2Real Transfer for Control of Cable-driven Robots via a Differentiable Physics Engine

通过可微分物理引擎控制电缆驱动机器人的 Real2Sim2Real 传输

• 发表时间: 2023
• 期刊: IEEE
• 作者: Wang, Kun;Johnson III, William R;Lu, Shiyang;Huang, Xiaonan;Booth, Joran;Kramer-Bottiglio, Rebecca;Aanjaneya, Mridul;Bekris, Kostas
• 通讯作者: Bekris, Kostas