Sensorless image based positioning control strategy design for AFM systems

As a typical tool for micro/nano imaging and manipulations, an atomic force microscope (AFM) has been widely used in a variety of areas including biology, materials, chemistry, physics, and so on. Specifically, to accurately implement nano-manipulation tasks for an AFM system, it is essential to design a high-precision positioning control strategy with nanometer precision. However, unfortunately, owing to the drawbacks of piezo-hysteresis, cross-coupling of piezoelectric and other nonlinear characteristics of this system, the positioning control of the probe for an AFM-based nanomanipulation system is still a challenging problem. Moreover, in most cases, a displacement sensor, which can detect position signals in real time, is hard to be applied into an AFM system directly, which leads to much more limitations in designing an efficient positioning method. To tackle these problems, in this paper, an image based high-precision positioning control strategy for AFM systems is first proposed, which presents high-precision probe positioning without using any displacement sensors, and instead, the scanned image is the only feedback information we used. In particular, firstly a constant velocity spiral scanning pattern as well as a Gaussian function-based distortion compensation model are proposed to obtain an undistorted image. Subsequently, a spiral trajectory planning based coarse positioning method is adopted to ensure a smooth and robust positioning process for most general samples. Then, considering some regular structure samples like particles, an improved local line scanning based method is proposed to further ensure more accurate positioning performance. Finally, some simulation results are presented to demonstrate the performance of the whole proposed positioning control strategy.