Control of the tight rope balancing robot

In this paper, a special type of underactuated robotic manipulator-tight rope balancing robotis-is analyzed, and the unique properties of this type of systems are revealed. The tight rope balancing robot is a two-link underactuated robot, which is a highly simplified model similar to a person balancing on a tight rope. In this work, control schemes are addressed from the views of both linear and nonlinear control theories. Simulations showing the motions of the robot using the control method developed in this paper are presented. These simulations verify the effectiveness of the control scheme.

I Introduction

The tight rope balancing robot is a highly simplified model of a tight rope walker. When an acrobat walks on a tight rope, a long rod is often used to keep balance. The robot considered here is a two-link manipulator, where the first link represents the torso of the acrobat, and the second link represents the balancing rod. Only the second joint of the robot is actuated while the first joint is free to rotate, so it falls into the class of underactuated manipulators. The difference between the tight rope balancing robot and the well-known acrobot [1]–[7] is that the center of gravity of the former coincides with the axis of joint two.