Global-Position Tracking Control for Highly Versatile Bipedal Robotic Walking

该项目将通过提高两足行走机器人的能力、安全性和可靠性，为提高国家健康、安全、福利做出贡献。双足机器人正在成为一项关键技术，用于广泛的重要应用，如搜索和救援，应急响应，家庭援助和医疗保健。该项目将解决在各种环境中控制和跟踪机器人运动的重要技术挑战。由此产生的多功能步行机器人将能够避开障碍物，例如在人口稠密的环境中避开行人，以及在灾难现场避开掉落的碎片。该项目将把研究与教育和推广活动结合起来，吸引来自不同群体的学生参与机器人和控制研究。该项目的目标是通过可靠的全局位置跟踪控制来提高混合系统的控制知识，以实现高度通用的两足机器人行走。该研究利用反馈控制理论和混合系统，为包括双足步行机器人在内的具有状态触发跳跃的通用混合系统创建了一种系统的跟踪控制器设计方法。为实现研究目标，将开展以下工作：(1)针对具有状态触发跳跃的混合系统的轨迹跟踪问题，提出一种新的参考轨迹修正方法，解决其中的间歇跟踪误差发散问题；(2)将李雅普诺夫理论扩展到具有状态触发跳跃的一般混合系统，以创建一类可证明能解决跟踪问题的控制器；(3)综合控制框架，利用所研究的控制器设计实现高通用性，同时具有可验证的稳定性、敏捷性和能效。该项目将为创造能够安全可靠地在现实世界中操作的下一代有腿机器人系统奠定基础。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 will contribute to the advancement of the national health, security and welfare, by improving capabilities, safety and reliability of bipedal walking robots. Bipedal robots are emerging as a critical technology for a broad range of important applications, such as search and rescue, emergency response, home assistance and health care. This project will address important technological challenges of controlling and tracking robot movements in various surroundings. The resulting highly versatile walking robot will be able to avoid obstacles, such as avoid pedestrians in human-populated environments and falling debris on disaster sites. This project will integrate the research with education and outreach activities for engaging students from diverse groups to participate in robotics and control research.The goal of this project is to advance the knowledge in the control of hybrid systems for achieving highly versatile bipedal robotic walking through reliable global-position tracking control. The research draws upon feedback control theory and hybrid systems to create a systematic method of tracking controller design for general hybrid systems with state-triggered jumps that include bipedal walking robots. To reach the research goal, the following objectives will be pursued: (1) to formulate a new method of reference trajectory modification for solving the intermittent tracking-error divergence issue underlying the trajectory tracking problem of hybrid systems with state-triggered jumps; (2) to extend Lyapunov theory to general hybrid systems with state-trigged jumps for creating a large class of controllers that can provably solve the tracking problem; and (3) to synthesize a control framework that exploits the researched controller design to achieve high versatility simultaneously with provable stability, agility, and energy efficiency for bipedal robotic walking. This project will lay a foundation for the creation of next-generation legged robot systems capable of safe and reliable real-world operations.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.