Humans have the amazing ability to walk with ease, navigating from daily environments to uneven and uncertain terrain with efficiency and robustness. If these same abilities could be imbued into robotic devices, the potential applications are far-reaching: from legged robots for space exploration to the next generation of prosthetic devices. Unfortunately the robotic bipedal walking community has been unable to agree on an appropriate model to generate anthropomorphic gait.
In this talk, I begin by describing how the sequence of contact point enforcements or discrete events along with a Lagrangian that is intrinsic to a biped completely determines the mathematical model for a biped. Given this insight, in the first part of the talk I describe a nine subject straight line walking motion capture experiment and two methods to extract temporal orderings: function fitting and persistent homology. Surprisingly the result of either method is that all participants regardless of age, height, sex, or weight had an identical temporal ordering of such events. In the second part of the talk, I describe how to generalize the detection of constraint enforcement by recasting the problem as an optimal control problem of switched dynamical systems. Given this result, we construct a numerical solution to the optimal control problem for a constrained switched nonlinear dynamical system with a running and final cost which provably converges to a local optimum.
Speaker Biography
Ram Vasudevan is a Ph.D. Candidate in Electrical Engineering and Computer Sciences at the University of California Berkeley. His research interests include hybrid systems, biologically inspired robotics, and computer vision. He is a Regent’s and Chancellor’s Scholar and a co-recipient of the Innovations in Networking Award presented by the Corporation for Education Network Initiatives in California. He received his B.S in Electrical Engineering and Computer Sciences and M.S. both from the University of California Berkeley in 2006 and 2009, respectively.