Online force control of a shape-memory-alloy-based 2 degree-of-freedom human finger via inverse model and proportional-integral-derivative compensator

In this work, a model-based controller is developed to track the force at fingertip of an artificial hand. To do so, shape-memory-alloy wires are implemented as an actuator in the finger. Besides, different aspects of modeling, including force relations, kinematics, and heat transfer analysis, are investigated. A modified version of Brinson's model is used to capture thermomechanical behavior of shape-memory-alloy wires. A controller is designed to control the applied potential difference between shape-memory-alloy wires and consequently control the electrical current in these wires based on the shape-memory-alloy wires model. The main goal of the proposed controller is force controlling of a 2-degree-of-freedom hand finger. This controller contains shape-memory-alloy constitutive model used for compensating system uncertainties. Furthermore, a proportional-integral-derivative controller/compensator is included in the closed-loop system. The compensator acts only on the derivative-type states, and this is one of the differences of this work compared to that of similar literature. The results of three arbitrary reference input signals are reported confirming the model prediction and simulation results are in good agreement with experimental tests. The analysis of the relative tracking error for an arbitrary reference signal is 11% in experimental test and 4% in the simulation.