Optimizing Dynamic Trajectories for Robustness to Disturbances Using Polytopic Projections

This paper focuses on robustness to disturbance forces and uncertain\npayloads. We present a novel formulation to optimize the robustness of dynamic\ntrajectories. A straightforward transcription of this formulation into a\nnonlinear programming problem is not tractable for state-of-the-art solvers,\nbut it is possible to overcome this complication by exploiting the structure\ninduced by the kinematics of the robot. The non-trivial transcription proposed\nallows trajectory optimization frameworks to converge to highly robust dynamic\nsolutions. We demonstrate the results of our approach using a quadruped robot\nequipped with a manipulator.\n