This paper presents the control of a two-degree-of-freedom piezoelectric actuator that exhibits hysteresis nonlinearity, creep nonlinearity, badly damped vibration, and cross couplings without using feedback sensors. The principle consists in compensating first the hysteresis, then the creep, and finally the vibration. The proposed compensation technique is multivariable and therefore is also able to reduce the cross couplings that are unwanted phenomena. The experimental tests demonstrate that the hysteresis that initially exceeds 19% is reduced to about 0.01%, while the creep is reduced from 5.5% to 0.04%. Regarding the vibration, the related overshoot that was initially 45% is completely removed. Note to Practitioners-This paper describes an approach to control and to automate dexterous precise positioning systems based on multiaxes piezoelectric actuators (PEAs). Two problems have motivated the approach investigated in this paper: 1) the presence of nonlinearities (hysteresis and creep), of badly damped vibration and of cross couplings in the multiaxes PEAs and 2) the lack of convenient sensors to feedback control them. Therefore, this paper proposes multivariable and complete feedforward control approach that does not require external sensors. This sensorless control architecture permits a low-cost and a high integration feature additionally to the fact that it is of great interest in applications at small scales where implementation of real-time measurement system is often difficult.

In this work we show that implementation of closed-loop control to silicon nanotweezers improves the sensitivity of the tool for mechanical characterization of biological molecules. Micromachined tweezers have already been used for the characterization of mechanical properties of DNA molecules [1] as well as for the sensing of enzymatic reactions on DNA bundle [2], [3]. However the resolution of the experiments does not allow the sensing on single molecules. Hereafter we show theoretically and experimentally that, reducing the resonance frequency of the system by the implementation of a state feedback, the sensitivity to stiffness variation is enhanced. Such improvement leads to better resolution for detection of enzymatic reactions on DNA [4].

This paper presents a study about the effects of environmental noise on millimeter sized grippers in cantilever configuration. The study is motivated and conducted aiming at assessing the level of accuracy loss when performing micromanipulation/microassembly tasks in noisy environments as well in typical microrobotics laboratories as in industrial locations or operating rooms. Ground motion and acoustic noises within a typical microrobotic laboratory are characterized in the frequency domain and their effects on cantilevers of different lengths are inspected. The relevance of a typical vibration isolation table is evaluated and the effects of low and high acoustic noises are assessed. A modeling of a cantilever with base excitation is thereafter conducted in the state space using finite difference formulation and a stabilization of a disturbed cantilever is obtained at the nanometer level in noisy environments allowing perspectives to high precision micromanipulation tasks in hostile locations.

MEMS-based microgrippers with integrated force sensor have proved their efficiency to perform dexterous micromanipulation tasks through gripping forces sensing and control. For force control, knowledge based models are more relevant and gives better physical significance than the use of black box models. However this approach is often limited by many problems commonly encountered in the MEMS (micro electromechanical systems) structures such as: complex architectures, nonlinear behaviors and parameters uncertainties due to fabrication process at the micrometer scale. For these reasons theoretical approaches must be compared with experiments. This paper describes a modelling approach of a MEMS-based microgripper with integrated force sensor while handling micro-glass balls of 80μm diameter. Therefore, a state space representation is developed to couple both the dynamics of the actuation and sensing subsystems of the gripper through the stiffness of the manipulated object. A knowledge based model is obtained for small displacements at the tip of the gripper arms (small gripping forces) and is compared with experimental approaches. Good agreements are observed allowing interesting perspectives for the control.