Hybrid Control for Agility and Efficiency in Bipedal Robots with Compliance

点击翻译按钮获取中文摘要

Proposal Number: 0600869PI - GrizzleIntellectual Merit: The objective of this research is to create a theory of hybrid systems that is rich enough to control a bipedal machine so that it can walk, run, and adapt its gait to varying terrain, like a human. A feedback design method will be developed that is systematic, provably correct, permits tradeoff analysis, and has effective symbolic and numerical tools for computations. The approach will be illustrated on legged machines with series compliant actuation. The hybrid zero dynamics of robots with compliant actuators will be studied and applied to specific motion control problems, such as energetically efficient running on a flat surface, and walking on uneven surfaces. Broader Impacts: There are important medical applications of bipedal locomotion research, including prostheses for the lower limbs and rehabilitation of walking and balance. Understanding stable dynamic motion in a machine is far simpler than in a human body, but it is a good place to start. Now that research is yielding an understanding of what it takes to achieve stable locomotion in machines capable of anthropomorphic gaits, it is possible to use this knowledge as a springboard in the search to recover locomotion ability in people who have suffered injuries. The PI is also cooperating with Thurston High School in S.E. Michigan to develop education material related to this research proposal that is appropriate for use in high school science courses.