Luis J. Ricalde

Chaos control and synchronization, with input saturation, via recurrent neural networks

This paper deals with the adaptive tracking problem of non-linear systems in presence of unknown parameters, unmodelled dynamics and input saturation. A high order recurrent neural network is used in order to identify the unknown system and a learning law is obtained using the Lyapunov methodology. Then a stabilizing control law for the reference tracking error dynamics is developed using the Lyapunov methodology and the Sontag control law. Tracking error boundedness is established as a function of a design parameter. The new approach is illustrated by examples of complex dynamical systems: chaos control and synchronization.

Inverse optimal nonlinear recurrent high order neural observer

This paper presents the design of an adaptive recurrent neural observer for nonlinear systems which model is assumed to be unknown. The neural observer is composed of a recurrent high order neural network which builds an online model of the unknown plant and a learning adaptation law for the neural network weights. This law is obtained by the Lyapunov methodology. The feedback law which guarantees stability of the estimation error is proved to be optimal with respect to a well defined cost functional.

Electrical Microgrid Optimization via a New Recurrent Neural Network

This paper presents the development and implementation of a new recurrent neural network for optimization as applied to optimal operation of an electrical microgrid, which is interconnected to the utility grid; moreover, it incorporates batteries, for energy storing and supplying, and an electric car. The proposed neural network determines the optimal amount of power over a time horizon of one week for wind, solar, and battery systems, including that of the electric car, in order to minimize the power acquired from the utility grid and to maximize the power supplied by the renewable energy sources. Simulation results illustrate that generation levels for each energy source over a time horizon can be reached in an optimal form.

DISCRETE-TIME REDUCED ORDER NEURAL OBSERVERS FOR UNCERTAIN NONLINEAR SYSTEMS

This paper focusses on a novel discrete-time reduced order neural observer for nonlinear systems, which model is assumed to be unknown. This neural observer is robust in presence of external and internal uncertainties. The proposed scheme is based on a discrete-time recurrent high order neural network (RHONN) trained with an extended Kalman filter (EKF)-based algorithm, using a parallel configuration. This work includes the stability proof of the estimation error on the basis of the Lyapunov approach; to illustrate the applicability, simulation results for a nonlinear oscillator are included.

RECURRENT NEURAL CONTROL FOR ROBOT TRAJECTORY TRACKING

This paper extends the results previously obtained for trajectory tracking of unknown plants using recurrent neural networks. The proposed controller structure is composed of a neural identifier and a control law defined by using the inverse optimal control approach, which has been improved so that less inputs than states are needed. The proposed new control scheme is applied to the control a robotic manipulator model.

Rollover prediction and control in heavy vehicles via recurrent neural networks

A state predictor is developed in order to estimate roll angle and lateral acceleration for tractor-semitrailers. Based on this prediction, an active control system is designed to prevent rollover. In order to develop this control structure, a high order recurrent neural network is used to model the unknown tractor semitrailer system; a learning law is obtained using the Lyapunov methodology. Then a control law, which stabilizes the reference tracking error dynamics, is developed using control Lyapunov functions. Via simulations, the control scheme is applied for speed-yaw rate trajectory tracking in a tractor-semitrailer during a cornering situation.

Optimal operation via a recurrent neural network of a wind-solar energy system

This paper focuses on the optimal operation of a wind-solar energy system, interconnected to the utility grid; moreover, it incorporates batteries for energy storing and supplying, and an electric car. It presents a neural network optimization approach combined with a multi-agent system (MAS). The objective is to determine the optimal amounts of power for wind, solar, and batteries, including the one of the electric car, in order to minimize the amount of energy to be provided by the utility grid. Simulation results illustrate that generation levels for each energy source can be reached in an optimal form using the proposed method.

Trajectory tracking via adaptive recurrent control with input saturation

This paper deals with the adaptive tracking problem for nonlinear systems in presence of unknown parameters, unmodelled dynamics and input saturation. A high order recurrent neural network is used in order to identify the unknown system and a learning law is obtained using the Lyapunov methodology. Then stabilizing control law for the reference tracking error dynamics is developed using the Lyapunov methodology and the Sontag control law. Tracking error boundedness is established as a function of a design parameter.

Inverse optimal design for trajectory tracking with input saturations via adaptive recurrent neural control

This paper is related to trajectory tracking problem for nonlinear systems, with unknown parameters, unmodelled dynamics and input saturations. A high order recurrent neural network is used in order to identify the unknown system and a learning law is obtained using the Lyapunov methodology. Then a control law, which stabilizes the tracking error dynamics, is developed using the inverse optimal control approach, recently introduced to nonlinear systems theory. Tracking error boundedness is established as a function of a design parameter. The applicability of the approach is illustrated via simulations, by synchronization of nonlinear oscillators.

Recurrent neural control for rollover prevention on heavy vehicles

An active control system is developed to prevent rollover in heavy vehicles. A high order recurrent neural network is used to model the unknown tractor semitrailer system; a learning law is obtained using the Lyapunov methodology. Then a control law, which stabilizes the reference tracking error dynamics, is developed using control Lyapunov functions. The control scheme is applied to the speed and speed-yaw rate trajectory tracking in a tractor-semitrailer during a cornering situation.