Tracking Control of Shape Memory Alloy Artificial Wrist Joint Based on Generalized Prandtl-Ishlinskii Inverse Model

Shape memory alloys (SMAs) are renowned for their exceptional properties. SMAs can undergo material crystal phase transitions in response to specific temperature variations, allowing them to regain their intended shape and overcome external stresses. However, the nonlinear hysteresis feature within the SMA actuators hinders accurate control of their behavior. In this paper, we propose an approach to construct an inverse hysteresis model based on a detailed examination of the principles underlying the phenomenological GPI model. Using the inverse model as a feedforward control mechanism, nonlinear plants can be approximately linearized through mathematical calculations, leading to improved controllability and superior performance. To further enhance control, we introduce a classical PID feedback controller as a complement to the GPI inverse model, forming a compound control methodology. In addition, we conduct displacement tracking experiments on an artificial wrist joint platform to gain a deeper understanding of the governing principles of SMAs.