Abstract:
Everyday human tasks such as picking up an object in one smooth motion, pushing a heavy door using the momentum of our bodies or pushing off a wall to quickly turn a corner involve complex dynamic interactions between the human and the environment, as well as switching dynamics when the robot makes and breaks contact. These dynamic interactions are critical in successful execution of such tasks. Thus, in order to enable robots to perform dynamic tasks effectively, we need to consider the dynamics of the robot, the individual objects in its environment, as well as the interactions between them.

In this talk, I will discuss our recent work on learning to perform dynamic manipulation tasks from expert demonstrations. We learn a switching linear dynamical model with contacts encoded in the switching conditions as a close approximation of our system dynamics. We then use discrete-time LQR as the differentiable policy class for data-efficient learning of control to develop a control strategy that operates over multiple dynamical modes and takes into account discontinuities due to contact. In addition to predicting interactions with the environment, our policy effectively reacts to inaccurate predictions such as unanticipated contacts. Through simulation and real world experiments, we demonstrate generalization of learned behaviors to different scenarios and robustness to model inaccuracies during execution.

Committee:
Prof. Oliver Kroemer (Chair)
Prof. Aaron Johnson
Prof. Zico Kolter
Cherie Ho