Wyniki wyszukiwania - BazTech

1

Design and experimentation of a self-tuning PID control applied to the 3DOF helicopter

Boubakir A, Labiod S., Boudjema F., Plestan F

Archives of Control Sciences

|

2013

|

Vol. 23, no. 3

311--331

EN

The paper presents design and experimental validation of a stable self-tuning PID controller for three degrees of freedom (3-DOF) helicopter. At first, it is proposed a self-tuned proportional-integral-derivative (PID) controller for a class of uncertain second order multiinput multi-output nonlinear dynamic systems to which the 3-DOF helicopter dynamic model belongs. Within this scheme, the PID controller is employed to approximate unknown ideal controller that can achieve control objectives. PID controller gains are the adjustable parameters and they are updated online with a stable adaptation mechanism designed to minimize the error between the unknown ideal controller and the used by PID controller. The stability analysis of the closed-loop system is performed using Lyapunov approach. It is proven that all signals in the closed-loop system are uniformly ultimately bounded. The proposed approach can be regarded as a simple and effective model-free control since the mathematical model of the system is assumed unknown. Experimental results are presented to verify the effectiveness of the proposed controller.

2

Hybrid adaptive control for speed regulation of an induction motor drive

Halbaoui K., Boudjema F., Boukhetala D.

Archives of Control Sciences

|

2009

|

Vol. 19, no. 2

151-164

EN

Model Reference Adaptive Control (MRAC) techniques may be used in high performance applications of nduction-motor (IM) drives to minimize adverse effects from changes in the load conditions and/or system parameters. Although the MRAC technique accounts for uncertainties and/or inaccuracies of the motor and load parameters in the design stage, its implementation on an nteger-based Digital Signal Processor (DSP) has several difficulties associated with the large dynamic range of the covariance matrix and the finite length of the DSP word. This paper investigates new form of a hybrid model reference adaptive speed control (HMRAC) to adapt the closed loop system including the plant with variation parameter to match with the reference model. The adaptive analog controller consists of a set of analog gain controller and a switching controller. The switching controller selects a controller from the set of analog controllers and connects it into the closed loop controlled system suitably. The duty rate of each analog controller will adapt the closed loop controlled system to be coincided with a reference model in reasonable sense. The importance of the hybrid controller is demonstrated by intensive experimental results. It is shown that the presented HMRAC for IM drive has fast tracking capability, smaller steady state error and is robust to load disturbance.

3

Nonlinear control design of active suspension based on full car model

Lagraa N., Boukhetala D., Boudjema F., Bloch G.

Archives of Control Sciences

|

2007

|

Vol. 17, no. 4

439-457

EN

This paper develops a nonlinear backstepping design scheme for the active control of vehicle suspension systems using a nonlinear full car model with seven degrees of freedom. The non-linear control law is designed to improve ride comfort and reduce the displacement and the acceleration of the sprung mass in the heaving, rolling, and pitching directions. To demonstrate the effectiveness of the proposed control strategy, several simulations are performed on the nonlinear system. They show that active suspension based on backstepping scheme gives superior performances over passive suspension.

4

Iterative learning control for robot manipulators

Bouakrif F., Boukhetala D., Boudjema F.

Archives of Control Sciences

|

2007

|

Vol. 17, no. 1

57-69

EN

In this paper, we present a time-domain iterative learning control scheme for the trajectory tracking problem of rigid robot manipulators that perform repeated tasks. The proposed control scheme comprises a computed torque control designed exploiting the approximated linear model of a manipulator and a learning law to compensate effects of nonlinear terms, that are ignored in obtaining the linear model, and the external disturbance. We show that the iterative learning controller is capable of effectively canceling the disturbances caused by nonlinear terms and other disturbance. The asymptotic stability of the closed-loop system is guaranteed, and the conditions of this stability are given. Simulation results on PUMA 560 robot show clearly efficiency of the proposed scheme.

5

Decentralized neural network controller for trajectory tracking of robot manipulators

Boukhetala D., Boudjema F., Djebiri M., Boucherit M.S., M'Sirdi N.K.

Archives of Electrical Engineering

|

2001

|

Vol. 50, nr 3

295-311

EN

This paper deals with the decentralized supervised control of robot manipulators using neural networks. First, a control law is synthesized by using the computed torque method. Since this latter is a model-based control technique, it presents some drawbacks such as sensitivity to parameter variations. To ovoid this problem, the control law is supervised using neural networks. Simulation results with robustness tests are provided to demonstrate whether the neural controller is capable to mimic the existing controller in one hand, and on the other hand, to compensate disturbances for which the existing controller did not show satisfactory results in terms of performances. A comparative study of both the decentralized structure and the centralized structure of the supervised control is also given.

PL

Artykuł dotyczy zdecentralizowanego nadzoru sterowania manipulatorów robotów przy zastosowaniu sieci neuronowych. Podano syntetyczny opis metody sterowania opartej o obliczenia momentu obrotowego. Posługuje się ona modelem wrażliwym na zmienność wprowadzanych parametrów. Aby ograniczyć ten wpływ sterowanie manipulatorów nadzorowane jest za pomocą sieci neuronowych. Podano wyniki symulacji oraz rezultaty prób odporności na zakłócenia dla zademonstrowania, czy sterownik neuronowy jest w stanie zastąpić zwykły sterownik, wykazując jednocześnie zdolność kompensacji zakłóceń, której zwykły sterownik nie posiada. Podano analizę porównawczą zdecentralizowanego oraz scentralizowanego podejścia do zagadnienia nadzoru sterowania.