Ing. Jiří Zemánek, Ph.D.

We introduce DICE (Discrete Integrated Circuit Electronics). Rather than separately develop chips, packages, boards, blades, and systems, DICE spans these scales in a direct-write process with the three-dimensional assembly of computational building blocks. We present DICE parts, discuss their assembly, programming, and design workflow, illustrate applications in machine learning and high performance computing, and project performance.

Rijke tube is a popular physical experiment demonstrating a spontaneous generation of sound in an open vertical pipe with a heat source. This laboratory instance of thermoacoustic instability has been researched due to its significance in practical thermoacoustic systems. We revisit the experiment, focusing on the possible use of the Rijke tube as a musical instrument. The novelty presented in this paper consists in shifting the goal from sound suppression to active sound generation. This called for modifying the previously investigated methods for stabilization of thermoacoustic oscillations into their excitation and control of their amplitude. On our way towards this goal, we developed a time-domain mathematical model that considers the nonlinear and time-varying aspects of the Rijke tube. The model extends the existing modeling and analysis approaches, which are mainly based in frequency domain. We also present an extension of the basic laboratory setup in the form of an array of Rijke tubes equipped with a single speaker used to control multiple Rijke tubes with different natural frequencies simultaneously.

In this paper, we present a novel approach to noncontact micromanipulation by controlled dielectrophoresis (DEP). To steer micro-objects in the desired way, the solutions reported in the literature use either DEP cages or amplitude modulation of the voltages applied to the electrodes. In contrast, we modulate the phases, that is, we control the phase shifts of the voltages applied to the electrodes, which simplifies the hardware implementation and extends the set of feasible forces. Furthermore, we introduce an innovative micro-electrode array layout, composed of four sectors with parallel (colinear) electrodes, which is capable of inducing an arbitrary movement in the manipulation area and is easy to fabricate using just an affordable one-layer technology. We then propose a closed-loop cascade control strategy based on real-time numerical optimization and deploy it to our experimental set-up. Numerical simulations and laboratory experiments demonstrate the manipulation capabilities such as positioning and steering of one or several microscopic objects (microspheres with a diameter of 50 μm) and even bringing two objects together and then separating them again. The results from simulations and experiments are compared and the positioning accuracy is evaluated in the whole manipulation area. The error in position is 8 μm in the worst case, which corresponds to 16% of the microsphere size or 0.7% of the manipulation range.

Ball and hoop system is a well-known model for the education of linear control systems. In this paper, we have a look at this system from another perspective and show that it is also suitable for demonstration of more advanced control techniques. In contrast to the standard use, we describe the dynamics of the system at full length; in addition to the mode where the ball rolls on the (outer) hoop we also consider the mode where the ball drops out of the hoop and enters a free-fall mode. Furthermore, we add another (inner) hoop in the center upon which the ball can land from the free-fall mode. This constitutes another mode of the hybrid description of the system. We present two challenging tasks for this model and show how they can be solved by trajectory generation and stabilization. We also describe how such a model can be built and experimentally verify the validity of our approach solving the proposed tasks.

Mathematical models of dielectrophoresis play an important role in the design of experiments, analysis of results, and even operation of some devices. In this paper, we test the accuracy of existing models in both simulations and laboratory experiments. We test the accuracy of the most common model that involves a point-dipole approximation of the induced field, when the small-particle assumption is broken. In simulations, comparisons against a model based on the Maxwell stress tensor show that even the point-dipole approximation provides good results for a large particle close to the electrodes. In addition, we study a refinement of the model offered by multipole approximations (quadrupole, and octupole). We also show that the voltages on the electrodes influence the error of the model because they affect the positions of the field nulls and the nulls of the higher-order derivatives. Experiments with a parallel electrode array and a polystyrene microbead reveal that the models predict the force with an error that cannot be eliminated even with the most accurate model. Nonetheless, it is acceptable for some purposes such as a model-based control system design.

The paper describes a simple time-optimal control strategy for a class of second-order bilinear systems with nonnegative inputs. The structure of the model is motivated by the task of noncontact manipulation of an object in a planar force field generated by a single source; such setup constitutes a basic building block for a planar manipulation by an array of force field sources. The nonnegative-in-control property means that an object (particle) placed freely in the field can only feel an attractive force towards the source. In this paper we further restrict the control inputs to a binary signal---the field can be switched on and off. The control objective is to bring the object to the origin (where the source of the force field is located) as fast as possible. The optimal switching strategy is proposed using geometric arguments and verified using numerical simulations and experiments with a laboratory platform for noncontact magnetic manipulation.

Various optical methods for measuring positions of micro-objects in 3D have been reported in the literature. Nevertheless, the majority of them are not suitable for real-time operation, which is needed, for example, for feedback position control. In this paper, we present a method for real-time estimation of the position of micro-objects in 3D1; the method is based on twin-beam illumination and requires only a very simple hardware setup whose essential part is a standard image sensor without any lens. The performance of the proposed method is tested during a micro-manipulation task in which the estimated position served as feedback for the controller. The experiments show that the estimate is accurate to within ∼3 μm in the lateral position and ∼7 μm in the axial distance with the refresh rate of 10 Hz. Although the experiments are done using spherical objects, the presented method could be modified to handle non-spherical objects as well.

The paper describes a novel control strategy for simultaneous manipulation of several microscale particles over a planar microelectrode array using dielectrophoresis. The approach is based on a combination of numerical nonlinear optimization, which gives a systematic computational procedure for finding the voltages applied to the individual electrodes, and exploitation of the intrinsic noise, which compensates for the loss of controllability when two identical particles are exposed to identical forces. Although interesting on its own, the proposed functionality can also be seen as a preliminary achievement in a quest for a technique for separation of two particles. The approach is tested experimentally with polystyrene beads (50 microns in diameter) immersed in deionized water on a flat microelectrode array with parallel electrodes. A digital camera and computer vision algorithm are used to measure the positions. Two distinguishing features of the proposed control strategy are that the range of motion is not limited to interelectrode gaps and that independent manipulation of several particles simultaneously is feasible even on a simple microelectrode array.

The paper presents application of dielectrophoresis for manipulation of a microparticle in a liquid medium above a planar surface. The force on the microparticle is exerted by a non-uniform AC electric field. The field is generated by a set of micro-electrodes patterned on the planar surface and connected to a multi-channel voltage waveform generator. An novel yet simple control approach based on changing the phase delays among voltages on the electrodes is introduced together with a description of an efficient hardware implementation. Laboratory experiments with a 250-micron polystyrene bead are documented. This work was partly motivated by the series of international competitions called Mobile Microrobotics Challenge and organized by NIST and taken over by IEEE RAS recently.

The prototype of a novel experimental platform for education and research on distributed manipulation by shaping physical force fields through arrays of sources will be presented. Description of both the hardware realization and (an example of) the control system implementation in Matlab/Simulink will be given. Some experimental results will be shown. The platform uses an array of coils to create and shape the magnetic field. One or several iron balls are placed into the field on a flat horizontal surface and they roll in response to the field. Positions of the balls are sensed in real time and these measurements can be used for closing the feedback loop commanding the current flowing through the coils. The controller can bring the objects to desired locations as well as steer them along predefined trajectories. Design of the platform is modular and allows for simple reconfiguring and expanding. Each module consists of four coils, driving and measuring electronics, a processor and connectors. Modules are interconnected by an RS-485 bus and commanded from PC. It is possible to control both the polarity and the magnitude of the current flowing though each coil. Position of the ball is measured using a resistive touch foil. Alternatively/additionally it can be obtained using image processing, which is a bit slower but allows for tracking of several balls. The mathematical model of the platform was built by experimental system identification while Matlab was used to gather the measured data and find the parameters of model that fit the measurements. Based on the model, a controller was designed. The controller keeps the popular cascade structure with the innermost loop responsible for achieving the forces requested by the higher-level control loops. Numerical optimization problem had to be solved in real time to find suitable setting of currents. Finally, a convenience of the use of iPad as a graphical user interface was demonstrated.

This report formulates the problem of a distributed planar manipulation realized by shaping a spatially continuous force field. It also suggests a control strategy based on feedback linearization. Force fields derived from potential fields are considered. The potentials are "shaped" by a set of spatially discrete "actuators" such as electrodes in the case of dielectrophoresis, electromagnets in the case of planar magnetic manipulators, or linear piezoelectric actuators in the case of deformable flat surfaces. The actuators form arrays. Distinguished feature of such force fields is that the contribution from an individual actuator usually affects the situation in the neighboring zones too, but usually not in too remote zones. As an idealization, the spatial domain is considered unbounded, which enables examination of asymptotic behavior of the manipulation scheme.

Communication in distributed control systems, especially those, that employ autonomous mobile devices moving in an unknown environment, requires high level of reliability, flexibility and speed in order to control individual nodes of the distributed network in real-time. This is the reason why several students' works at the Department of Control Engineering involved development of mobile appliances with autonomous behaviour, which differ in terms of timing demands and amount of transmitted data, thus allowing choice of different types of wired as well as wireless communication. Structure of these mobile appliances as well as their communication with superior stationary node in the distributed network is described in this paper.

The paper gives a short overview of the authors' recent research activities at the intersection of microfluidics, electrokinetics and self-assembly. The particular research introduced in the paper aims at developing an electric-field-assisted self-assembly procedure for meso- and micro- (and possibly nano-) scale systems. The goal is to steer tiny particles such as LED chips, thick film resistors or microbeads submerged in a shallow liquid pool around by controlling electric voltage applied to an electrode array at the bottom of the pool. Dielectrophoresis as the major tool is introduced first, a brief report on numerical simulations using a commercially available FEM solver is then given, design and fabrication of the microfluidic chamber featuring microelectrode array is described in some detail, and finally laboratory experiments are presented. As an outcome of the experiments, a series of videos demonstrating various responses of particles were captured.