Cascade control of robotic fingers with anthropomorphic inspiration

This paper presents the design of cascade controller for robotic fingers designed based on an anthropomorphic inspiration. These fingers are driven by Pneumatic Artificial Muscle actuators. The high nonlinear dynamics of these actuators and the inherent hysteresis in their behavior lead to the modelling and control problems that cause a lack of robustness in the hand's performance. The actuator has been mathematically modelled as a nonlinear second order system and the estimator of the system uncertainty has been incorporated into adaptive backstepping control law. The cascade controller is designed by integrating the adaptive backstepping controller and PID controller for position control of the robotic fingers. The experiment results have proven that the proposed controller is capable to compensate for the coulomb friction force which is the system uncertainty and improves the position control of the robotic fingers. In addition, the robotic fingers have introduced an adaptive grasping for cylindrical-shaped objects with different diameters. The robotic hand has imitated the human hand in terms of size, weight and grasping.