Controlling an SMA-based Mini Robot by Emotional Learning

The prevalence of miniature robots in the past decades proves their advantages over large-scale robotics. This paper presents the design and simulation-based control of a mini-robot comprising a shape memory alloy (SMA) actuator to control the rotary position of a plate attached to the system. The modeling equations for the SMA wire are presented and employed in simulating the SMA behavior. The developed controller is an emotional learning controller based on fuzzy concept that can learn the system’s behavior while following the desired trajectories. The controller modifies its parameters to reduce the final tracking error to precisely reach the target position with partial state feedback.