Geometric Control for Dynamic Aerial Manipulation and Transportation

动态空中操纵和运输的几何控制

• 批准号: 1538869
• 负责人: Koushil Sreenath
• 金额: $ 35.69万
• 依托单位: Carnegie-Mellon University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2015
• 资助国家: 美国
• 起止时间: 2015-09-01 至 2018-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1538869&HistoricalAwards=false
• 关键词: Geometric; Control; Dynamic; Aerial; Manipulation

There is a need for safe and reliable aerial load transport through complex environments. In developing countries road networks may not exist or may be inaccessible for significant parts of the year due to seasonal rain, flood, or snow. In dense urban areas, energy and time costs of transportation are rapidly increasing due to road congestion. Autonomous aerial vehicles have emerged as a fast, economical alternative to delivery using traditional ground-based infrastructure. This project will show how the cargo-carrying limitations of single aerial vehicles may be overcome using multiple vehicles to cooperatively transport cable-supported loads. Current control approaches impede progress in this domain by ignoring the tight dynamical coupling between multiple aerial robots and the shared cable-suspended load, the dynamics on manifolds, the unilateral cable tension constraints, the collision constraints, the multiple hybrid dynamical modes, and the physical system limits. This project will address these issues, and furthermore show how operational flexibility associated with redundant vehicles and the hybrid nature of cable support may be exploited to achieve greater maneuverability of the load-carrying formation.The overarching goal of this work is to understand the science at the intersection of geometric control and convex optimization for achieving cooperative highly-dynamic aerial manipulation that explicitly addresses the hybrid dynamics of the system while providing formal guarantees of stability and safety. The transformative nature of this research stems from its ability to generate theoretical advances in feedback control on multiple levels, while also being firmly grounded in being demonstrated on a physical system with stringent performance requirements and safety constraints. This work is therefore guided by the following goals (1) Formulate constrained geometric control for systems evolving on manifolds to enforce input and state constraints; (2) Develop safety-critical geometric control for systems evolving on manifolds to enforce formal collision-free dynamic motion; and (3) Demonstrate highly dynamic multi-agent cooperative aerial manipulation and transportation that switches between multiple hybrid dynamical modes. As a result of these research goals, this work has the potential to enable the next generation of aerial load transportation using teams of small unmanned aerial robots.

需要通过复杂环境的安全可靠的空中载荷运输。在发展中国家，由于季节性降雨、洪水或降雪，道路网可能不存在或在一年中的大部分时间无法通行。在人口密集的城市地区，由于道路拥堵，交通的能源和时间成本迅速增加。自主飞行器已成为使用传统地面基础设施交付的快速、经济的替代方案。该项目将展示如何使用多个车辆合作运输电缆支撑的负载来克服单个飞行器的货物运输限制。目前的控制方法阻碍了在这一领域的进展，忽略了多个空中机器人和共享的电缆悬挂负载，流形上的动态，单向电缆张力约束，碰撞约束，多个混合动力学模式之间的紧密动力学耦合，和物理系统的限制。本项目将解决这些问题，并进一步展示了如何利用与冗余飞行器相关的操作灵活性和缆索支撑的混合性质来实现承载编队的更大机动性。这项工作的首要目标是了解几何控制和凸优化交叉点的科学，以实现协同的高度-动态空中操纵，明确解决了系统的混合动力学，同时提供稳定性和安全性的正式保证。 这项研究的变革性源于其在多个层面上产生反馈控制理论进步的能力，同时也坚定地立足于在具有严格性能要求和安全约束的物理系统上进行演示。 因此，这项工作是由以下目标指导：（1）制定约束几何控制流形上的系统发展，以加强输入和状态约束;（2）开发安全关键的几何控制流形上的系统发展，以加强正式的无碰撞动态运动;（3）演示高动态多智能体合作空中操纵和运输，切换多个混合动力模式。 由于这些研究目标，这项工作有可能使下一代的空中载荷运输使用小型无人驾驶航空机器人的团队。