This paper describes a method for feedforward tracking control of linear and non-linear systems with periodic desired trajectories. The method utilizes adaptive Finite Impulse Response (FIR) filters to realize an adaptive inverse control scheme. Compared to previously reported frequency domain methods, this technique can be implemented in realtime and generates a coefficient update every sampling instance. The proposed method is successfully applied to the adaptive inverse control of an experimental nanopositioning system. The maximum RMS error during both large-range and high-speed scans was 0.23%. This is comparable to previously reported frequency-domain techniques and is far superior to the performance achievable with standard feedback methods.
American Control Conference, 2009 (ACC '09). Proceedings of the American Control Conference 2009 (St Louis, MO 10-12 June, 2009) p. 1676-1681