Adaptive Sliding-Mode Control of a Charged Particle in an Ion Trap

Currently, commercial ion traps used for mass spectrometry are controlled in an open-loop manner with sinusoidally varying inputs. In this paper, we discuss the possible advantages of adding nonlinear feedback control to this system and demonstrate them through numerical simulations. Using sliding-mode control, we find that we can have a particle fall onto a trapping surface of our choosing, despite the presence of uncertainty in the system. In addition, when used in an open-loop fashion, the sliding-mode input creates stable attractors in the phase space. This shows that nonsinusoidal periodic inputs can effectively trap a group of particles. When an adaptive component is added to the closed-loop sliding-mode controller, we see that a simulated particle of unknown mass and charge can be successfully trapped and driven onto a desired surface. In addition, if that trajectory satisfies the persistent-excitation condition, then the controller can attain perfect estimation of the unknown parameters, thus measuring the particle mass and charge without ejecting it from the trap. These simulation results suggest a number of interesting experiments.