INSPIRE Track 1: The Mathematics of Balance in Mechanical Systems with Impacts, Unilateral Constraints, Underactuation and Hyper-sensing: Application to Agile bipedal Locomotion

本次INSPIRE奖的部分资金来自工程局电气、通信和网络系统部的能源、电力和自适应系统计划，土木工程、机械和制造创新部的控制系统计划，工程局的信息和智能系统部的稳健智能计划，数学和物理科学局的数学科学部的应用数学计划。虽然我们正在见证一场名副其实的两足机器人建造技术革命，但反馈控制算法中体现的平衡科学仍处于初级阶段，这些算法允许这些机器人站立、行走和跨越障碍物。目前的反馈算法需要针对每台机器和每种步态进行费力、耗时的反复调整。如果两足机器人要发挥其潜在的作用，从帮助家里的老人或体弱者到灾难和救援响应，它们的控制软件开发过程必须建立在更好的科学基础上。智力优势：该项目利用力学、MEMS设备、数学和反馈系统，使两足机器人的建模和反馈控制设计实践发生了翻天覆地的变化。这项工作旨在促进对腿部运动的科学理解，以便对性能目标进行数学测试，以确定可靠模型的可行性，并为基于模型的反馈控制器设计提供原则性方法。该项目的理论成果正在密歇根大学的3D两足机器人上进行实验评估。广泛的影响：两足运动研究的重要医学应用包括双腿假肢和用于恢复受伤后行走和平衡的设备。PI利用他们在数学、传感、反馈系统和机器人方面的跨学科工作来促进STEM学科的发展。研究人员积极参与快速传播，包括在YouTube频道上发布的视频、电视节目，以及每周参观他们的机器人实验室。

This INSPIRE award is partially funded by Energy,Power and Adaptive Systems Program in the Division of Electrical, Communications and Cyber Systems in the Directorate For Engineering, Control Systems Program in the Division of Civil, Mechanical, and Manufacturing Innovations, in the Directorate for Engineering, Robust Intelligence Program in the Division of Information and Intelligent Systems in the Directorate for Computer and Information Science and Engineering, Applied Mathematics Program in the Division of Mathematical Sciences in the Directorate for Mathematical and Physical Sciences. While we are witnessing a veritable revolution in bipedal robot construction technology, the science of balance, as embodied in the feedback control algorithms that allow these robots to stand, walk, and step over obstacles is still in its infancy. Current feedback algorithms demand laborious, time consuming trial-and-error tuning specific to each machine and each gait. If bipedal robots are ever to fulfill their potential roles from assisting the elderly or infirm in their homes to disaster and rescue response, their control software development process must be founded on better science. Intellectual Merit: The project draws upon mechanics, MEMS devices, mathematics, and feedback systems to make a sea change in the practice of modeling and feedback control design in bipedal robots. The work seeks to advance scientific understanding of legged locomotion so that performance objectives can be tested mathematically for feasibility on reliable models and principled methods exist for model-based feedback controller design. The theoretical results of the project are being evaluated experimentally on a 3D bipedal robot at the University of Michigan.Broader Impacts: Important medical applications of bipedal locomotion research include lower-limb prostheses and devices for the rehabilitation of walking and balance after injury. The PIs use their interdisciplinary work in mathematics, sensing, feedback systems, and robotics to promote STEM subjects. The researchers are actively involved in rapid dissemination including videos posted on YouTube channels, television programs, and weekly tour of their robotics laboratory.