A stochastic state prediction in AFM based nanomanipulation

The reliability of AFM based nanomanipulation is one of the prerequisites for nano fabricating, thus uncertainties of the target nano-particle position need to be controlled within a pointed limits, which lead to two problem to be solved: one is how to describe state uncertainties at every different sample times; the other is how to deduce the next state uncertainty according to the current one. This paper provides a method to construct a stochastic predictive model to covering those problems. In the concrete, the predictive model uses Gaussian and Beta distribution as well as Monte Carlo Method to describe and deduce the state uncertainties respectively. The method is also applied to the process of pushing a nano-particle using AFM.