We havepreviously developed amobile robot position controller basedonbackstepping control algorithm. In thispaper,we proposetheextention ofmentioned controller withanaimtorapidly decrease thecontrol torques neededtoachive thedesired position andorientation of mobile robot. Theparameters ofthis controller areadjusted bygenetic algorithm. Thesamegenetic algorithm wasused forevolution ofthecontrol parameters ofamultivariable PI velocity controller described witha fullmatrix. The performance oftheproposed systemisinvestigated using a dynamic modelofanonholonomic mobile robot withfriction. Simulation results showthegoodquality ofbothvelocity and position tracking capabilities ofamobile robot. 1~~~~~~. EUROCON2005

Abstract In this paper, we propose two level control system for a mobile robot. The first level subsystem deals with the control of the linear and angular volocities using a multivariable PI controller described with a full matrix. The position control of the mobile robot represents the second level control, which is nonlinear. The nonlinear control design is implemented by a modified backstepping algorithm whose parameters are adjusted by a genetic algorithm, which is a robust nonlinear optimization method. The performance of the proposed system is investigated using a dynamic model of a nonholonomic mobile robot with friction. We present a new dynamic model in which the angular velocities of wheels are main variables. Simulation results show the good quality of position tracking capabilities a mobile robot with the various viscous friction torques.

Abstract This paper proposes an extension of neural network identification capabilities for on-line identification of a nonlinear closed-loop control system. The neural network (NN) is trained on-line using the backpropagation optimization algorithm with an adaptive learning rate. The optimization algorithm is performed at each sample time to compute the optimal control input. The results confirm the effectiveness of the proposed neural network based identification scheme and control architecture.

This work is considering three significant factors that affect blood glucose level: food intake, hereditary predisposition and stress. Goal of this paper is to observe blood sugar level in human organism as a dynamic MISO (Multi Input, Single Output) system, and to describe it with differential equations and control system blocks. The system has three inputs; food (carbohydrates), hereditary factor and stress, and a single output--blood glucose level. Basically, several logical assumptions have been made, as the result of few medical researches. A model that gives outputs, very similar to real ones (measurements of glucose level in human body) is used for more detailed analysis. This model is very suitable for computer simulations and it can easily be tested for different input arrangements. Using this property of the system, several modes of food consumption have been proposed, in order to retain blood sugar level inside recommended limits.

In this paper, Pade's rational functions have been simulated for approximating several characteristic values of time delay regarding the plant time constant. Several representative plants were tested in order to show in which cases Pade’s function approximates time-delay block well. Only if the ratio of time delay versus time constant of the plant is rather great, or the plant contains emphasized numerator dynamics; approximation capabilities get poorer. The convergence rate of n-order Pade’s function has been also analyzed by using Taylor series and phase-frequency characteristics.