Title:  Whole-body Control of Wheeled Inverted Pendulum Humanoids

Committee:

Dr. Seth Hutchinson, ECE, Chair , Advisor

Dr. Evangelos Theodorou, AE

Dr. Henrik Christensen, CoC

Dr. Justin Romberg, ECE

Dr. Byron Boots, CoC

Abstract:

We propose to develop a framework for controlling a Wheeled Inverted Pendulum (WIP) Humanoid to perform useful interactions with the environment while dynamically balancing itself on two wheels. As humanoid platforms are characterized by several degrees of freedom, they have the ability to perform several tasks simultaneously, such as balancing themselves, maintaining a specific body pose, controlling the gaze, lifting a load or carrying a tray of cups filled to the brim. These tasks are all performed simultaneously, and the whole body participates in achieving each objective, with priorities assigned to each. The control also has to operate within constraints of angle and torque limits on each joint, as well as safety constraints of avoiding self-collision and collision with obstacles. This problem is referred as Whole-Body Control in the wider humanoid literature, and several successful solutions have recently been demonstrated for bipedal humanoid platforms. Our focus in this work is to develop a framework for whole-body control of WIP humanoids that can be applied directly on the physical robot, which means that it can be made robust to modeling errors as well as able to incorporate constraints on control and state as mentioned above. 

The proposed approach is hierarchical with a low level controller responsible for controlling the manipulator/body and a high-level controller that defines center of mass (CoM) targets for the low-level controller to control zero dynamics of the system driving the wheels. The low-level controller plans for shorter horizons while considering more complete dynamics of the system, while the high-level controller plans for longer horizon based on an approximate model of the robot for computational efficiency. Our core contributions are

  - Showing how to isolate the dynamics of the manipulator from those of the wheels such that the resulting model is amenable to existing whole-body control techniques, such as operational space control and quadratic programs (QP)

  - Using differential dynamic programming (DDP) to generate optimal trajectories for center of mass to control wheel motion