Carlos E. Castañeda

Discrete-Time Neural Sliding-Mode Block Control for a DC Motor With Controlled Flux

An adaptive discrete-time tracking controller for a direct current motor with controlled excitation flux is presented. A recurrent neural network is used to identify the plant model; this neural identifier is trained with an extended Kalman filter algorithm. Then, the discrete-time block-control and sliding-mode techniques are used to develop the trajectory tracking. This paper also includes the respective stability analysis for the whole closed-loop system. The effectiveness of the proposed control scheme is verified via real-time implementation.

Decentralized neural identifier and control for nonlinear systems based on extended Kalman filter

A time-varying learning algorithm for recurrent high order neural network in order to identify and control nonlinear systems which integrates the use of a statistical framework is proposed. The learning algorithm is based in the extended Kalman filter, where the associated state and measurement noises covariance matrices are composed by the coupled variance between the plant states. The formulation allows identification of interactions associate between plant state and the neural convergence. Furthermore, a sliding window-based method for dynamical modeling of nonstationary systems is presented to improve the neural identification in the proposed methodology. The efficiency and accuracy of the proposed method is assessed to a five degree of freedom (DOF) robot manipulator where based on the time-varying neural identifier model, the decentralized discrete-time block control and sliding mode techniques are used to design independent controllers and develop the trajectory tracking for each DOF.

Direct current motor control based on high order neural networks using stochastic estimation

An adaptive discrete-time tracking controller for a direct current (DC) motor with controlled excitation flux is presented. A high order neural network in discrete-time is used to identify the plant model; this network is trained with an extended Kalman filter where the associated state and measurement noises discrete-time covariance matrices are calculated with stochastic estimation. Then, the discrete-time block control and sliding mode techniques are used to develop the trajectory tracking for the angular position of a DC motor with separate winding excitation. Numerical computation presented in this paper shows that the proposed method provides accurate estimation for the covariance matrices associated in the extended Kalman filter.

Real-time torque control using discrete-time recurrent high-order neural networks

This paper presents a discrete-time direct current (DC) motor torque tracking controller, based on a recurrent high-order neural network to identify the plant model. In order to train the neural identifier, the extended Kalman filter (EKF) based training algorithm is used. The neural identifier is in series-parallel configuration that constitutes a well approximation method of the real plant by the neural identifier. Using the neural identifier structure that is in the nonlinear controllable form, the block control (BC) combined with sliding modes (SM) control techniques in discrete-time are applied. The BC technique is used to design a nonlinear sliding manifold such that the resulting sliding mode dynamics are described by a desired linear system. For the SM control technique, the equivalent control law is used in order to the plant output tracks a reference signal. For reducing the effect of unknown terms, it is proposed a specific desired dynamics for the sliding variables. The control problem is solved by the indirect approach, where an appropriate neural network (NN) identification model is selected; the NN parameters (synaptic weights) are adjusted according to a specific adaptive law (EKF), such that the response of the NN identifier approximates the response of the real plant for the same input. Then, based on the designed NN identifier a stabilizing or reference tracking controller is proposed (BC combined with SM). The proposed neural identifier and control applicability are illustrated by torque trajectory tracking for a DC motor with separate winding excitation via real-time implementation.

Continuous-time neural control for a 2 DOF vertical robot manipulator

This paper presents a continuous-time neural control scheme for identification and control of a two degrees of freedom (DOF) direct drive vertical robot manipulator model, on which effects due to friction and gravitational forces are both considered. A recurrent high-order neural network (RHONN) structure is proposed in order to identify the plant model to then, based on this neural structure, derive a neural controller using the backstepping design methodology. The trajectory tracking performance of the neural controller is illustrated via simulations results, which suggest the validity of the proposed approach for its implementation in real-time.

Stochastic Control of a Quadrotor

A stochastic controller for a quad rotor via state feedback approach is designed. The extended Kalman filtering (EKF) algorithm is implemented in order to generate the estimated states. Simulation results show that the proposed scheme achieves stabilization and trajectory description for the quad rotor.

Real-time torque control for a DC motor using recurrent high order neural networks

This paper presents a discrete-time direct current (DC) motor torque tracking controller, based on a recurrent high order neural network (RHONN) to identify the plant model. Using this model, a control law is derived, which combines block control and sliding modes techniques. The applicability of the scheme is illustrated via real time implementation for a DC motor with separate winding excitation.

Real-Time Implementation of a Neural Integrator Backstepping Control via Recurrent Wavelet First Order Neural Network

Wavelets are designed to have compact support in both time and frequency, giving them the ability to represent a signal in the two-dimensional time–frequency plane. The Gaussian, the Mexican hat, and the Morlet wavelets are crude wavelets that can be used only in continuous decomposition. The Morlet wavelet is complex-valued and suitable for feature extraction using continuous wavelet transform. Continuous wavelets are favoured when a high temporal resolution is required at all scales. In this paper, considering the properties from the Morlet wavelet and based on the structure of a recurrent high-order neural network model, a novel wavelet neural network structure, here called recurrent wavelet first-order neural network, is proposed in order to achieve a better identification of the behavior of dynamic systems. The effectiveness of our proposal is explored through the design of a centralized neural integrator backstepping control scheme for a two degree-of-freedom robot manipulator evolving in the vertical plane. The performance of the overall neural identification and control scheme is verified through numerical simulation using the mathematical model for a benchmark prototype. Moreover, real-time results validate the effectiveness of our proposal when using a robotic arm, of our own design, powered by industrial servomotors.

Neural Block Control via Integrator Backstepping for a Robotic Arm in Real-Time

This paper presents an online neural identification and control scheme in continuous-time for trajectory tracking of a robotic arm evolving in the vertical plane. A recurrent high-order neural network (RHONN) structure in a block strict-feedback form is proposed to identify online in a series-parallel configuration, using the filtered error learning law, the dynamics of the plant. Based on the RHONN identifier structure, a stabilizing controller is derived via integrator backstepping procedure. The performance of the neural control scheme proposed is tested on a two degrees of freedom robotic arm, of our own design and unknown parameters, powered by industrial servomotors.

Linear Time-Varying Luenberger Observer Applied to Diabetes

We present a linear time-varying Luenberger observer (LTVLO) using compartmental models to estimate the unmeasurable states in patients with type 1 diabetes. The proposed LTVLO is based on the linearization in an operation point of the virtual patient (VP), where a linear time-varying system is obtained. LTVLO gains are obtained by the selection of the asymptotic eigenvalues where the observability matrix is assured. The estimation of the unmeasurable variables is done using Ackermann’s methodology. The Lyapunov approach is used to prove the stability of the time-varying proposal. In order to evaluate the proposed methodology, we designed three experiments: 1) VP obtained with Bergman’s minimal model; 2) VP obtained with Hovorka’s model; and 3) real patient data set. For both experiments 1) and 2), it is applied a meal plan to the VP, where the dynamic response of each state model is compared with the response of each variable of the time-varying observer. Once the observer is obtained in experiment 2), the proposal is applied to experiment 3) with data extracted from real patients, and the unmeasurable state space variables are obtained with the LTVLO. LTVLO methodology has the feature of being updated each time instant to estimate the states under a known structure. The results are obtained using simulation with MATLAB and