Robust Motion Trajectories for an Uncertain Flexible Robotic System Using Ensemble Control

Abstract

Abstract. Controlling one or more flexible components connected to a maneuvering robotic structure has been a long-standing problem in the robotics community. Typical applications include the control of flexible beams, robotic manipulators, spacecraft systems, and cranes. A large number of different approaches for solving this problem have been proposed, e.g., phase-plane trajectory analysis, switching property analysis, Pontryagin’s maximization principle, and various input shaping methods. Many of the studies have used a canonical spring–mass–damper system as a proxy for the practical plant. The previous approaches offer limited robustness in the presence of uncertainty in the flexible modes, which are generally challenging to model accurately. This article presents a new approach based on the concept of ensemble control to improve the robustness of motion control for flexible systems. In particular, rest-to-rest time-optimal slewing control of a planar structure is studied. Compared to available standard methods, the proposed method is observed to offer superior performance. Several examples are presented to illustrate the efficacy of the approach.