The position control of magnetic medical microrobots is influenced by several environmental uncertainties including the unknown characteristics of the medium and the imaging precision. Furthermore, measuring the physical attributes of the microrobots is a challenging issue. To provide a model-free position control approach for the magnetic medical microrobots, a saturation-tolerant Adaptive Fuzzy Sliding-Mode Control (AFSMC) is designed in this study. In the proposed approach, each control input comprises a fuzzy inference term utilized to approximate an unknown nonlinear function including uncertain forces, a robust part derived to compensate for the fuzzy approximation error and disturbances, and a compensating gain for input saturation. By utilizing the second theorem of Lyapunov and Barbalat’s lemma, it is proved that the closed-loop system is asymptotically stable. The effectuality of the presented controller is assessed by means of two experimental scenarios. The results show that the magnitudes of the tracking errors corresponding to a spiral reference path are less than 0.2 mm at the end of the motion. Moreover, in the test conducted in a 3D printed Aorta phantom, the minimum and the maximum values of the tracking errors are 1.22 mm and 0.95 mm, respectively. —This article introduces an innovative, model-free technique specifically designed to tackle the complex challenges of position control in magnetic medical microrobots. Achieving precise control over these microrobots is a challenging task, compounded by the complexity of accurately measuring their physical properties and the characteristics of their surrounding medium. This challenge is further exacerbated by the issue of input saturation, which can compromise system stability. Our pioneering control method is designed to navigate these obstacles effectively. It functions under the assumption that both the lower and upper saturation limits are unknown, and it eliminates the necessity to model the forces acting on the microrobot. Experimental results confirm the method’s effectiveness in accurately tracking various reference trajectories. These findings suggest that our method holds significant promise for various medical applications. IEEE
