Robotic fish is an interesting and prospective subject to develop. The simplest fish swimming mode to be mimicked for
fish robots is the ostraciiform mode which only requires caudal fin flapping. An almost submerged ostraciiform fish
robot was constructed to study its swimming characteristics. The swimming direction can be controlled by changing the
mean angle of caudal fin oscillation. Experiments were conducted to study the behavior of the fish robot and in particular,
the transfer function between swimming path angular rate and mean angle of the caudal fin oscillation were identified.
Error to signal ratio quantity was used to determine how well the model fits with the experimental data. This
identification model was used to design a 2-degree-of-freedom PID controller that meets some specific requirements to
improve the steering performance.
A smart material is known to be able to generate large force in broad bandwidth in a compact size. However it needs
relatively large voltage to drive it and this makes the system bulky. In this paper, first, we introduce an improved version
of miniaturized piezo actuator driver and modeling of the dynamics of the piezo actuator, LIPCA. ARX model was used
to model the dynamics of the LIPCA. We applied rectangular waves to the LIPCA and measured its responses with a
strain gauge and a signal processing circuit. A 5th order model was obtained from the input/output data and applying
identification algorithm. Secondly, we designed a simple PID controller based on the obtained model to improve the
characteristics of the LIPCA actuator.
A self oscillation loop in vibrating gyroscope based on the phase locked loop (PLL) was proposed and a phase error in the PLL was analyzed. The self oscillation loop is a nonlinear feedback loop, which keeps a self-generated and sustained oscillation. In vibrating type gyroscope, a structure needs a driving oscillation at resonant frequency, which is provided by the self oscillation loop. In order to sustain the loop to be stable oscillation and to operate at the precise resonant frequency of the gyroscope, phase locking condition is essentially needed. Phase locked loop is a suitable component for such a purpose. In oscillation loop, PLL provides exact phase shift to oscillate at the resonant frequency point over wide frequency range. However, the performance of the PLL is affected by certain condition of the circuit, e.g. phase error due to the variation of the component values and noise-sensitive features. This directly leads the performance degradation of the gyroscope. Therefore, phase error analysis of PLL was performed and the robust phase shifting circuit was newly suggested. The experiments were accomplished to verify the suggested method.
This paper presents a novel approach for demodulation method of signal processing circuit in MEMS gyroscope. Since the MEMS gyroscope utilizes Coriolis acceleration that produces a modulated signal of the input angular velocity and driving signal, in order to measure the original angular rate, the demodulation process is essentially needed. The conventional AM demodulation process in MEMS gyroscope is sensitive to the phase difference between the output signal and the modulation reference signal. Moreover, the output is easily affected by nonlinear and noisy properties of a multiplying circuit. Proposed method eliminates the phase tuning of the demodulation stage and the multiplying process of the signal processing circuit that are likely to be major error factors of signal processing circuit but are essential parts of the conventional demodulation process in MEMS gyroscope. The proposed method utilized the envelope detection scheme of AM demodulation in communication system and modified it to apply to the electromechanical system of gyroscope. Experiments were accomplished to verify the performances. From the results, the proposed method shows a satisfactory performance without a multiplying component and tuning effort of the phase in signal process circuit.
This paper describes the development of biomimetic structure systems with LIPCA (LIghtweight Piezo-Composite Actuator) and battery supported power control unit. To apply LIPCA as a biomimetic actuator for the control surface of small unmanned air vehicle, a battery supported power control unit was developed, which is composed of a lithium polymer, one step-up converter, four power switching high voltage transistors, on Schmitt triggered comparator, and control logics. A simple RC circuit is used to sample the voltage applied to the LIPCA. H-switch was applied which is composed of the four high voltage transistors to control the voltage or charge and its polarity applied to the LIPCA. From experiments, it was observed that the developed biomimetic adaptronic systems could be constructed with relatively compact and light units and could produce enough displacement and force to be used as a control surface for the elevator and the rudder of a small unmanned vehicle.