Robust motion control synthesis with essential dynamics for hybrid-driven continuum robots

Abstract

Incorporating system dynamics into the design of centralized control architectures is increasingly recognized as an effective approach to improve the motion control of continuum robots. However, the huge computational demands of most existing dynamic models pose significant challenges for their application in real-time control scenarios. In this paper, for hybrid-driven continuum robots, we first develop two simplified dynamic models that capture the essential dynamic characteristics while ensuring high computational efficiency. Subsequently, by incorporating the simplified models into feedback control, a robust dual-loop control framework suitable for real-time applications is presented. Specifically, the inner loop adopts these simplified models to counteract the inherent nonlinear dynamics of systems, thereby achieving a linear input/output relationship. The outer loop focuses on stabilizing the entire closed-loop system. Moreover, an extra robust term is designed and incorporated into the outer loop to mitigate the effects induced by modeling errors. Comparative experiments based on a hybrid-driven continuum robot with two segments are implemented to validate the effectiveness of these simplified models and their control synthesis.