Motion Planning with Limit-Cycle Gait Primitives
Abstract-Much of the current research on dynamic legged robots focuses on the design and control of mechanisms that ultimately will have the capacity to generate and sustain motions that mimic the prowess, agility and endurance of animals. Undoubtedly, reproducing features of animal locomotion on machines is a task of formidable complexity. Yet, to realize the potential of such machines in real-world applications, basic movement patterns must be coordinated and regulated to synthesize more complex behaviors that achieve desirable task-level objectives. This talk presents a hierarchically consistent framework for synthesizing locomotion control and motion planning strategies capable of translating descending task-level commands to suitable low-level control actions that harness the platform’s locomotion capabilities. We adopt a formal approach that relies on the construction of a library of locomotion primitives in the form of suitably parameterized limit cycle gaits that capture the platform’s locomotion capabilities; this is where questions about platform operation, stability maneuverability and energetics can be addressed. Composition of these primitive behaviors according to task-level commands then results in a wide range of periodic and non-periodic agile locomotion behaviors that afford formal safety guarantees and can faithfully realize high-level planning objectives.
Prof. Poulakakis earned his Ph.D. in Electrical Engineering from the University of Michigan, Ann Arbor, MI, in 2008 and served as a Postdoctoral Research Associate at Princeton University, Princeton, NJ, before joining the University of Delaware, Newark, DE, in 2010, where he currently serves as an Associate Professor of Mechanical Engineering. His research interests are in the area of dynamics and control with application to bio-inspired robotic systems, specifically legged robots. In 2014 he received a Faculty Early Career Development Award from the National Science Foundation to investigate task planning and motion control for legged robots at different scales.