Zdzisław Gosiewski

The Robust Control of Magnetic Bearings for Rotating Machinery

The fast progress in the applications of active magnetic suspension systems needs to apply the modern control theory. This paper deals with H∞ and H2 control of rigid rotor movement, which is supported in magnetic bearings. The robust control of magnetic bearings is investigated analytically. The nominal model of active magnetic suspension of rotor and the uncertainty model were derived. The standard PID control and robust control are compared and performance of nominal feedback configuration with weights is presented. We propose a robust control with a multi-objective controller to achieve good robust stability when the model of a plant is uncertain. The behavior of multiplicative uncertainty of magnetic suspension system is shown. The aim of optimal robust control is to improve the magnetic suspension taking into account the energy limitation (i.e., to avoid the saturation of actuators). The H2 performance and H∞ performance depend on a proper selection of weighting functions. So a very important step in the controller design process is to choose the appropriate weight functions: We, Wu, Wd. The influence of noise is limited by weight functions. We also put limits on input and output signals. The stability of a system with disturbance interaction is discussed. The simulations of a well-posed and internally stable magnetic system are presented. The success of the robust control is demonstrated through results of numerical simulations.

Vision-based obstacle avoidance for unmanned aerial vehicles

A problem of unmanned aerial vehicles (UAVs) being able to avoid obstacles and collisions by means of processing the images taken by one camera is presented in the paper. The optical flow method based on gradient method of Lukas-Kanade was used to obtain the tested structure in 3D space environment (extraction of images depth). This method allows to generate local information flow patterns from particular neighborhood of surrounding points. Presented algorithm for obstacle location and estimation of its shape was developed and implemented on the workstation with real time uClinux system. Proposed vision-based obstacle avoidance algorithm is demonstrated in simulation and in hardware in the loop flight tests on a fixed-wing UAV.

Kalman filter realization for orientation and position estimation on dedicated processor

Abstract This paper presents Kalman filter design which has been programmed and evaluated in dedicated STM32 platform. The main aim of the work performed was to achieve proper estimation of attitude and position signals which could be further used in unmanned aeri-al vehicle autopilots. Inertial measurement unit and GPS receiver have been used as measurement devices in order to achieve needed raw sensor data. Results of Kalman filter estimation were recorded for signals measurements and compared with raw data. Position actualization frequency was increased from 1 Hz which is characteristic to GPS receivers, to values close to 50 Hz. Furthermore it is shown how Kalman filter deals with GPS accuracy decreases and magnetometer measurement noise.

Formation Flight Control Scheme for Unmanned Aerial Vehicles

Cooperative control of Unmanned Aerial Vehicles is currently being researched for a wide range of applications. Applicability of aerial unmanned systems might be increased by formation flight. In this paper, we present an application of a proportional-integral controller for formation flight of three Unmanned Aerial Vehicles (UAVs). A decentralized cooperative control scheme is used with information flow modeled by a leader-follower structure. Control laws, based on velocity and position errors between the vehicles, are defined and derived for each of the three flying objects in formation. The main goal of this work is a simulation study. Three-dimensional motion of the three-object formation was performed and analyzed using a six-degrees-of-freedom state space quadrotor model. The simulation studies presented allow verifying the adopted control structure. Convergence time is used as an indicator of the control system quality and the formation stability. The presented research and diagram of control algorithms tests were done before experimental tests and field trials of formation flight.

Modelling of Dynamic and Control of Six-Rotor Autonomous Unmanned Aerial Vehicle

The paper describes matters of modeling and control of aerial vehicle in rotorcraft configuration. Equations of motion were derived and dynamic model of six-rotor was build. To find the best adjustment of model to real object the dynamics of the drives was joined to the control plant model. Trimming and linearization were performed. Open loop system and LQR controller were checked in simulation environment. Next the faults of the actuator/sensor elements were arranged and the minimal number of observed outputs and drives for LQR stabilization process were specified.

The Differential Passive Magnetic Bearing for High-Speed Flexible Rotor

Lab stand of a differential passive magnetic bearing is presented in the paper. The passive bearing will be used in high-speed flexible rotor. The one neodymium magnet MP 41x15x10 – N38 and two neodymium magnets MP 40x22x10 – N38 were used to design the passive magnetic bearing. The mathematical model of differential bearing, laboratory model and distribution of magnetic flux density are presented and analyzed.

Modeling of Beam as Control Plane for a Vibration Control System

An active vibration damping system with the use of piezoelements fixed to the structure is presented in the paper. The optimal location of the piezo-stripes on the structure is calculated for given cost functions. Coefficient correlations between control forces, strains, mode shapes, frequency changes, modal masses and modal stiffness are analyzed in order to find simpler method for calculation of the quasi-optimal localization of piezo-sensors and piezo-actuators. Reduced model of the open-loop system was identified and the simulations were carried out in order to find the best controller to reduce the amplitude of vibrations. The correctness of the selection of the controller was verified and investigated experimentally.

Reduction Methods of Structure Models for Control System Purposes

To design vibration control system for flexible structures their mathematical model should be reduced. In the paper we consider the influence of the model reduction on the dynamics of the real closed-loop system. A simply cantilever beam is an object of consideration since we are able to formulate the exact analytical model of such structure. As a result of reduction the model with low frequency resonances is usually separated from the high frequency dynamics because high frequency part of the model is naturally strong damped. In order to estimate dynamical system for control purposes in the paper we applied a few orthogonal methods such as: modal, Rayleigh-Ritz and Schur decompositions. As it is shown all methods well calculate resonances frequencies but generate different anti-resonances frequencies. From control strategy in point of view of the flexible structures these anti-resonances have significantly influence on the stability and dynamics of the closed-loop systems.

Simulation Model of an Electromagnetic Multi-Coil Launcher for Micro Aerial Vehicles

The paper presents a model of the 10-coil launcher, which can be used as a catapult of micro aerial vehicles (MAVs). The main advantage of such a device is the possibility of controlling its acceleration. We have also shortly described main disadvantages of pneumatic and rubber launchers and compared them to our solution which involved magnetic phenomena. The history and the classification of magnetic launchers were also mentioned. The investigated magnetic system was described in details. The system voltage equations and energetic conversions which take place during the launch were presented. With the use of finite element method we calculated the lumped parameters of the system (self-and mutual inductances). Nonlinear model was implemented in the MATLAB packet. Launch operation was divided into an acceleration and a braking phase. Thus, nine coils operated as accelerators and the last one was employed as a magnetic brake. The system was controlled by means of PID regulators which were tuned manually. Current values in the modelled coils were also limited by bang-bang controllers. In the last part of the paper we presented results of simulations and discussion about possible improvements of the model in the future.

An FPGA Implementation of the Robust Controller for the Active Magnetic Bearing System

The article presents the project design flow that leads to the implementation of the robust controller law in the FPGA chip. The designed robust FPGA controller is going to be used in the heteropolar active magnetic bearing system. The hardware and software architectures of the designed controller are presented. The hardware consists of the market available Spartan-3 Development Board and two specially designed A/D and D/A converters boards. The software architecture is made of several VHDL entities that are translated into the target FPGA chip. The results of the experimental preliminary tests show that the dynamic properties of the designed controller are very good and the authors hope that the dynamic performance, especially the stability of the whole active magnetic bearing system, will improve.