Santiago Torres

A self-tuning neuromorphic controller: application to the crane problem

Abstract This paper is concerned with the design and application of a self-tuning controller, aided by means of neural network s (NN). The structure of the controller is based on the use of neural networks as an implicit self-tuner for the controller. The aim o f this approach is to take advantage of the learning properties of the neural networks to increase the performance of the self-tuning. The a pplication of this technique is performed on an overhead crane. The control objective is to suppress undesirable oscillations during op eration of the crane. First, some simulations were carried out, as well as a comparison with a standard self-tuning method, that demon strate the advantages of this method. After this, a real-time implementation on a scale prototype of a crane was done to verify th e applicability of the method.

Adaptive computer control of anesthesia in humans

This paper presents an efficient computer control technique for regulation of anesthesia in humans. The anesthetic used is propofol and the objective is to control the degree of hypnosis of the patient. The paper describes the basic hardware/software setup of the system and the closed-loop methodologies. The bispectral index (BIS) is considered as the feedback signal. The control methods proposed here are based in the use of proportional integral controllers with dead-time compensation to avoid undesirable oscillations in the BIS signal during the process. The compensation is based on the Smith predictor. To guarantee the applicability of the method to different patients, an adaptive module to tune the compensator is developed. Some real and simulated results are presented in this work to attest the efficiency of the methods used.

An Application of a Neural Self-Tuning Controller to an Overhead Crane

A neural network-based self-tuning controller is presented. The scheme of the controller is based on using a multilayer perceptron, or a set of them, as a self-tuner for a controller. The method proposed has the advantage that it is not necessary to use a combined structure of identification and decision, common in a standard self-tuning controller. The paper explains the algorithm for a general case, and then a specific application on a nonlinear plant is presented. The plant is an overhead crane which involves an interesting control problem related to the oscillations of the load mass. The method proposed is tested by simulation in different conditions. A comparison was made with a conventional controller to evaluate the efficiency of the algorithm.

On the Design and Implementation of a Neuromorphic Self-Tuning Controller

This paper deals with the design and implementation of a neural network-based self-tuning controller. The structure of the controller is based on using a neural network, or a set of them, as a self-tuner for a controller. The intention of this approach is to take advantage of the ability to learn of the neural networks and to use them in place of an identifier in the conventional self-tuner scheme. The work is divided into two main parts. The first one is dedicated to the design of the self-controller. And the second is an application of the algorithm on a nonlinear system: an overhead crane. Some simulations were carried out to verify the efficiency of the self-tuner and then a real-time implementation on a scale prototype was performed.

A predictive control algorithm with interpolation for a robot manipulator with constraints

The paper presents an efficient control algorithm applied on a two-link robot manipulator with input constraints. The algorithm proposes a suboptimal solution to the predictive control problem with infinite prediction horizon, by mean of interpolations between the unconstrained optimal solution and other constrained solutions. The control strategy is based on inserting the predictive controller in an adaptive perturbation scheme. The efficiency of the proposed strategy is shown by simulation.

Model-based controller for anesthesia automation

This paper presents an approach to anesthesia control using a model-based controller. General anesthesia with propofol is considered. The proposal tries to take advantage of the benefits of model-based controllers to improve the performance of control in anesthesia. The controller proposed is based on the application of two control actions. First, a nominal term is applied obtained from the inverse dynamics model. This action is corrected by adding a second term that compensates modeling errors, disturbances, etc. To compute the correction, a linearization of the model is considered around the nominal state and optimization is performed to compute the control action. Several results obtained in simulation are presented to test the efficiency of the method.

Disturbances rejection on a robot arm using an efficient predictive controller

Abstract This work proposes a predictive controller with interpolation in order to improve the behaviour of a typical MPC when the system presents constraints. Particularly, it is interesting to see how the interpolation, between the solution of the optimal unconstrained problem and other feasible solutions, assures the stability of the system in presence of disturbances. The system in which the controller is applied is a two-link robot manipulator arm. The predictive controller is inserted in an adaptive perturbation scheme to change adequately the nominal inputs, given by an inverse dynamics controller, in order to reject the disturbances produced. The efficiency of the proposed strategy is shown by simulation.

Seguimiento de Trayectorias en Robots Manipuladores: Revisión de Soluciones y Nuevas Propuestas

The tracking problem in robot manipulators has been afforded by applying a great variety of controllers, since easy designs based on PD controllers until complex control designs based on adaptive and robust algorithms. These last techniques show some drawbacks, i.e., some bounds in the robot dynamics have to be considered or the controller does not afford with the system constraints. This work makes a revision of the existing classical control techniques for manipulators and proposes a new set of robust and predictive controllers in order to avoid the mentioned problems. Particularly, a self-adaptive robust controller is described which avoids the error produced by an inexact cancellation of the nonlinear dynamics terms. This controller is improved by means of predictive algorithms that include the robot constraints in the control law. This work includes real and simulation results of a PUMA-560 arm of Unimation, which prove the satisfactory performance of the proposed controllers.

FOUR ROTOR HELICOPTER CONTROL LABORATORY PLANT

Abstract In this paper, a test-bed for teaching in multivariable system is presented. Firstly, the different aspects of the prototype construction will be described, making a special emphasis in the mechanics, the design of the sensorial and actuation systems and the prototype control. Next, the real time control software will be explained. The mathematical model of the plant is presented in order to design a control strategy, test it in simulation and validate in the real system.

Efficient Control of Robot Manipulators with Model Disturbances

The aim of this work is to present a new adaptive strategy for articulated robotic systems in order to reject adequately the tracking errors due to disturbances in the model. The technique is based on a well known robust controller for robotic manipulators proposed by Spong, which is derived using Lyapunovs direct method. The adaptation law acts in an essential parameter of the robust controller, the uncertainty bound parameter, in order to improve its performance. This paper presents real-time experimental results with a Puma-560 robot, comparing the performance of the new strategy with the classical PD controller with gravity compensation and with Spongs robust controller, in case of model uncertainties. It is shown that the adaptive controller improves the performance of previous control strategies