F.M. Raimondi

Least squares and genetic algorithms for parameter identification of induction motors

Abstract This paper deals with off-line parameter identification of induction motors by means of least square (LS) techniques and genetic algorithms (GA), using stator voltages, stator currents and velocity as input–output data. For analytical identification by LS algorithms, filtering of experimental data is performed by means of anticausal filters. Two models useful for identification are derived in which the products of acceleration and rotor fluxes, usually neglected, are taken into account. The GA-based identification method consists of the determination of the best parameters which match input–output behaviour of the motor. Both methods are investigated and compared by means of experiments carried out on a 1-kW induction motor.

Trajectory tracking of underactuated underwater vehicles

This paper deals with a control strategies for underactuated underwater vehicles whose target is the tracking of a space trajectory. A cascade control strategy is employed which brings to a control law consisting of: 1) a kinematic control law, derived from the vehicle kinematic model, which forces this model to track the reference trajectory; and 2) a dynamic control law which forces the system to track the reference signals given by the kinematic control law. Conditions for asymptotic tracking of the trajectory are given with reference to the standard dynamical model of the above vehicle. An observer of the marine current is also added in order to process the control law. Simulation tests illustrate the proposed approach.

An adaptive control law for robotic manipulator without velocity feedback

In this paper, a new adaptive control law is designed for robotic manipulators, based on the use of reference velocities instead of the actual ones and feedback signals generated from position errors. The law in question is suitable for trajectory tracking and positioning tasks. Its peculiarities are implementation without velocity measurements and estimation, high signal-to-noise ratio in control torques and absence of parameter drift in positioning tasks. Experimental tests are shown with the aim to confirm the validity of the control law and to illustrate its actual effects on the behaviour of the system.

Identification of nonlinear systems described by Hammerstein models

This paper deals with a method for identification of nonlinear systems suitable to be described by Hammerstein models consisting of a static nonlinearity followed by an ARX linear model. The estimation of the static nonlinearity is carried out supplying the system with a sequence of step signals of various amplitude and determining the corresponding steady-state responses. The estimation of the parameters of the ARX linear system is carried out by means of a least square estimator using data generated supplying the system with a Pseudorandom Binary Sequence (PRBS). The method in question is able to identify static nonlinearities of general type, also with hysteresis and/or discontinuities. Simulation results confirm the validity of the method and its capabilities with respect to some other methods.

A new method for optimal synthesis of wavelet-based neural networks suitable for identification purposes

Abstract This paper deals with a new method for optimal synthesis of Wavelet-Based Neural Networks (WBNN) suitable for identification purposes. The method uses a genetic algorithm (GA) combined with a steepest descent technique and least square techniques for both optimal selection of the structure of the WBNN and its training. The method is applied for designing a predictor for a chaotic temporal series

Adaptive and Robust Techniques for Friction Compensation in Motion Control of Robotic Manipulators

Abstract This paper deals with the study of friction compensation methods for motion control of robotic manipulators. First of all, modelling of the friction is carried out decomposing it into three components, Coulombs, viscous and stiction components. Then, an adaptive model-based control law is designed which includes an adaptive compensation scheme of the modelled friction. Finally, a variable structure robust control loop is designed whose inclusion in an adaptive control scheme designed using a simplified model of the friction allows to obtain better tracking accuracy. Some results of experimental tests, carried out on a two links SCARA manipulator, are displayed with the aim to show the effectiveness of the proposed compensation methods.

Identification of a Hammerstein model for DC/DC converters operating in CCM

This paper deals with a method for identification of a Hammerstein model of DC-DC converters operating in continuous conduction mode (CCM). This model has the duty cycle and the output voltage as input and output, respectively; it consists of a static nonlinearity and a linear and timeinvariant model. The aim of the modeling the system by means of a Hammerstein model is due to its capability of describing the converter in a range of steady-state operating points instead of a desired well defined operating point as occurs for the small-signal models which are the more common mathematical description to approach the study of the converters themselves. The nonlinear characteristic of the Hammerstein model is constructed by determining ii certain number of its input-output couples using the results of measures carried out at the steady-state for constant inputs. The linear model is identified using results of a transient corresponding to a suitable two-values PRBS sequence whose values are chosen among the above input-output couples of the nonlinearity. The identification of the Hammerstein model is carried out using experiments carried out in PSpice environment.

Fuzzy Data Fusion for Real-World Mapping Using 360° Rotating Ultrasonic Sensor

Abstract Mobile robot perception of the external environment is limited by the features of the used sensor. An useful technique used to improve robot perception is data fusion. This paper presents an approach to build a map of an unknown environment applying fuzzy data fusion methods to data acquired through an ultrasonic sensor. Conditioning of these data and motion control of the mobil robot by fuzzy data fusion are also described. The resulting two dimensional map is used for path planning and navigation. The proposed approach is exrperimentally tested using real distance measures acquired by a 360° rotating sensor.

ROTOR FLUX OPTIMAL ESTIMATION FOR INDUCTION MOTOR CONTROL

Abstract The aim of this paper is to analyze and design reduced order observers of the rotor flux of induction motors. The design requirements are: a) the convergence rate of the rotor flux estimation error; b) a low sensitivity to stator and rotor resistance variations; c) a low sensitivity to errors due to the implementation of the observers on microprocessor-based systems. It is shown that, in order to satisfy the requirements a)-c), it is sufficient to solve a constrained optimization problem according to a criterion in which these requirements appear explicitly. The implementation of the observer is discussed. The observer is tested by simulation and experiments.

A New Method of Velocity Estimation Based on Variable Temporal Basis Using Incremental Encoder

Abstract This paper deals with analysis and synthesis of algorithms for digital conditioning of signals generated by incremental encoders to estimate velocity of rotating devices for control purposes. Main objectives are to obtain high accuracy at low and high velocity and low tracking delays during accelerations. A digital conditioning method is described, Which uses a polynomial of order n whose coefficients are updated so as to fit the n+1 most recent velocity data acquired on a variable temporal basis. Digital sinlulations and experimental findings are shown with the ainl to validate the proposed estimation method and compare it with other methods.