Projects

@inproceedings{Ragazzon2019,
title = {Generalized Lyapunov Demodulator for Amplitude and Phase Estimation by the Internal Model Principle},
author = {M. R. P. Ragazzon and S. Messineo and J. T. Gravdahl and D. M. Harcombe and M. G. Ruppert},
year = {2019},
date = {2019-09-04},
booktitle = {8th IFAC Symposium on Mechatronic Systems},
abstract = {Effective demodulation of amplitude and phase is a requirement in a wide array of applications. Recent efforts have increased the demodulation performance, in particular, the Lyapunov demodulator allows bandwidths up to the carrier frequency of the signal. However, being inherently restricted to a single order filtering of the input signal, signal components outside its passband are not sufficiently attenuated for all applications, such as in multifrequency atomic force microscopy. In this article, the structure of the Lyapunov demodulator is transformed to an equivalent form, taking advantage of the internal model representation of the sinusoid to be tracked. A generalization of this formulation allows the application of standard filtering techniques in order to shape the characteristics of the demodulator, while retaining the perfect tracking condition provided by the internal model. Guidelines for the filter design are provided in order to achieve the desired characteristics, such as filtering order, tracking bandwidth, and transient performance. The resulting generalized Lyapunov demodulator structure is highly flexible, allows for direct employment of any standard filter type, is computationally simple, and easy to implement requiring only a bandpass filter, a single integrator, and two nonlinear transformations. Numerical results demonstrate the effectiveness of the approach, and provide a comparison of the various filters considered.},
note = {accepted for publication},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}

@article{Yong2019,
title = {Design and Analysis of Low-Distortion Demodulators for Modulated Sensors},
author = { S. I. Moore and M. G. Ruppert and D. M. Harcombe and A. J. Fleming and Y. K. Yong },
url = {https://ieeexplore.ieee.org/document/8762188},
doi = {10.1109/TMECH.2019.2928592},
year = {2019},
date = {2019-07-17},
journal = {IEEE/ASME Transactions on Mechatronics},
volume = {Early Access},
abstract = {System-based demodulators in the form of a Kalman and Lyapunov filter have been demonstrated to significantly outperform traditional demodulators, such as the lock-in amplifier, in bandwidth sensitive applications, for example high-speed atomic force microscopy. Building on their closed loop architecture, this article describes a broader class of high-speed closed-loop demodulators. The generic structure provides greater flexibility to independently control the bandwidth and sensitivity to out-of-band frequencies. A linear time-invariant description is derived which allows the utilization of linear control theory to design the demodulator. Experimental results on a nanopositioner with capacitive sensors demonstrate the realization of arbitrary demodulator dynamics while achieving excellent noise rejection.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}

@inproceedings{C19d,
title = {Multivariable Model-less Feedforward Control of a Monolithic Nanopositioning Stage With FIR Filter Inversion},
author = {M. Omidbeike and A. A. Eielsen and Y. K. Yong and A. J. Fleming },
year = {2019},
date = {2019-07-02},
booktitle = {International Conference on Manipulation, Automation and Robotics at Small Scales (MARSS)},
address = {Helsinki, Finland},
abstract = {A model-less approach for inversion of the dynamics of multivariable systems using FIR filters is described. Inversion-based feedforward techniques have been widely used in the literature to achieve high-performance output tracking. The foremost difficulties associated with plant inversions are model uncertainties and non-minimum phase zeros. Various model-based methods have been proposed to exclude nonminimum phase zeros when inverting both single-input and single-output (SISO), and multiple-input and multiple-output (MIMO) systems. However, these methods increase the model uncertainty as they are no longer exact. To overcome these difficulties a model-less approach using FIR filters is presented. The results when applying the feedforward FIR filter to a multivariable nanopositioning system is presented, and they demonstrate the effectiveness of the feedforward technique in reducing the cross-coupling and achieving significantly improved output tracking.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}

@inproceedings{omidbeike2019axis},
title = {A Five-Axis Monolithic Nanopositioning Stage Constructed from a Bimorph Piezoelectric Sheet},
author = {M. Omidbeike and Y. K. Yong and S. I. Moore and A. J. Fleming
},
year = {2019},
date = {2019-07-02},
journal = {Int. Conference on Manipulation, Automation and Robotics at Small Scales (MARSS)},
address = {Helsinki, Finland},
abstract = {The paper describes design, modeling and control of a five-axis monolithic nanopositioning stage constructed from a bimorph piezoelectric sheet. In this design, actuators are created by removing parts of the sheet using ultrasonic machining. The constructed nanopositioner is ultra-compact with a thickness of 1 mm. It has a X and Y travel range of 15.5 µm and 13.2 µm respectively; a Z travel range of 26 µm; and a rotational motion about the X-and Y-axis of 600 µrad and 884 µrad respectively. The first resonance frequency occurs at 883 Hz in the Z-axis, and the second and third resonance frequency appears at 1850 Hz, rotating about the X-and Y-axis. A decentralized control strategy is implemented to track Z, θx and θy motions. The controller provides good tracking and significantly reduces cross-coupling motions among the three degrees-of-freedom.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}

@inproceedings{Ruppert2019,
title = {Model-based Q Factor Control for Photothermally Excited Microcantilevers},
author = {M. G. Ruppert and B. S. Routley and A. J. Fleming and Y. K. Yong and G. E. Fantner},
year = {2019},
date = {2019-07-01},
booktitle = {Int. Conference on Manipulation, Automation and Robotics at Small Scales (MARSS)},
address = {Helsinki, Finland},
abstract = {Photothermal excitation of the cantilever for dynamic atomic force microscopy (AFM) modes is an attractive actuation method as it provides clean cantilever actuation leading to well-defined frequency responses. Unlike conventional piezo-acoustic excitation of the cantilever, it allows for model-based quality (Q) factor control in order to increase the cantilever tracking bandwidth for tapping-mode AFM or to reduce resonant ringing for high-speed photothermal offresonance tapping (PORT) in ambient conditions. In this work, we present system identification, controller design and experimental results on controlling the Q factor of a photothermally driven cantilever. The work is expected to lay the groundwork for future implementations for high-speed PORT imaging in ambient conditions.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}

@inproceedings{Harcombe2019,
title = {Modeling and Noise Analysis of a Microcantilever-based Mass Sensor},
author = {D. M. Harcombe and M. G. Ruppert and A. J. Fleming},
year = {2019},
date = {2019-07-01},
booktitle = {Int. Conference on Manipulation, Automation and Robotics at Small Scales (MARSS)},
address = {Helsinki, Finland},
abstract = {Nanomechanical devices have the potential for practical applications as mass sensors. In microcantilever based sensing, resonance frequency shifts are tracked by a
phase-locked loop (PLL) in-order to monitor mass adsorption. A major challenge in minimizing the mass detection limit comes from the noise present in the system due to thermal, sensor and oscillator noise. There is numerical difficulty in simulating PLLs, as both low frequency phase estimates and high frequency mixing products need to be captured resulting in a stiff problem. By using linear system-theoretic modeling an in-depth analysis of the system is able to be conducted overcoming this issue. This provides insight into individual noise source propagation, dominant noise sources and possible ways to reduce their effects. The developed model is verified in simulation against the non-linear PLL, with each achieving low picogram sensitivity for a 100 Hz loop bandwidth and realistically modeled noise sources.},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}