Direct inverse linearization of piezoelectric actuator's initial loading curve and its applications in full-field optical coherence tomography (FF-OCT)

Piezoelectric actuator (PEA) has become popular in spectroscopic optical coherence tomography (SOCT) as it can provide depth dependent spectra with sub?nanometer positioning resolution. However, the inherent hysteresis of the PEA introduces unwanted distortions to the obtained spectra and image. OCT imaging can be completed during the first forward motion of the PEA, i.e., the initial loading curve (ILC). Therefore, this paper proposes a feedforward hysteresis compensation method for the linearization of the ILC, aiming at improving the spectroscopic and imaging accuracy of OCT without any feedback on the PEA's displacement or feedback controllers. A polynomial operator is adopted as the hysteresis compensator and the coefficients are directly identified from the measured ILC, following the direct inverse modeling approach. Experimental results on a standalone PEA verifies that the proposed method is superior to the popular Prandtl-Ishlinskii model in the linearization of the ILC. This proposed method is further implemented on a full?field OCT system to linearize the motion of the PEA?driven scanner. Experimental results show that the ILC of the scanner can be efficiently linearized. In vivo skin imaging is then performed to evaluate the performance of the proposed method. Ultra-high precision linearization enables remarkable positioning accuracy in the skin constituents and skin layers observed in the in vivo forearm cross-sectional skin image. Depth-dependent spectral analysis could quantify molecular/chemical composition of the tissue constituents accurately. ? 2020 Elsevier Ltd