Controlling Pivoting Gait Using Graph Model Predictive Control

Pivoting gait is efficient for manipulating a big and heavy object with relatively small force, in which robot iteratively tilts the object, rotates it around vertex, then puts down to floor. However, pivoting can easily fail even external disturbance due its instability nature. To cope this problem, we propose controller robustly control motion during by introducing two modes, i.e., one double-support mode, manipulate light faster speed, other quadruple-support lower speed. gait, graph model predictive applied taking into account of these modes. By adaptively switching mode according disturbance, stably perform if object.