Open Access. Powered by Scholars. Published by Universities.®
- Publication Year
Articles 31 - 41 of 41
Full-Text Articles in Engineering
Infinite Time Optimal Neuro Control For Distributed Parameter Systems, S. N. Balakrishnan, Radhakant Padhi
Infinite Time Optimal Neuro Control For Distributed Parameter Systems, S. N. Balakrishnan, Radhakant Padhi
Mechanical and Aerospace Engineering Faculty Research & Creative Works
The conventional dynamic programming methodology for the solution of optimal control, despite having many desirable features, is severely restricted by its computational requirements. However, in recent times, an alternate formulation, known as the adaptive-critic synthesis, has given it a new perspective. In this paper, we have attempted to use the philosophy of adaptive-critic design to the optimal control of distributed parameter systems. An important contribution of this study is the derivation of the necessary conditions of optimality for distributed parameter systems, described in discrete domain, following the principle of approximate dynamic programming. Then the derived necessary conditions of optimality are …
Robust Adaptive Critic Based Neurocontrollers For Systems With Input Uncertainties, S. N. Balakrishnan, Zhongwu Huang
Robust Adaptive Critic Based Neurocontrollers For Systems With Input Uncertainties, S. N. Balakrishnan, Zhongwu Huang
Mechanical and Aerospace Engineering Faculty Research & Creative Works
A two-neural network approach to solving optimal control problems is described in this study. This approach called the adaptive critic method consists of two neural networks: one is called the supervisor or critic, and the other is called an action network or controller. The inputs to both these networks are the current states of the system to be controlled. Each network is trained through an output of the other network and the conditions for optimal control. When their outputs are mutually consistent, the controller network output is optimal. The optimality is limited to the underlying model. Hence, we develop a …
Frequency Domain Robustness Analysis Of Hopfield And Modified Hopfield Neural Networks, Jie Shen, S. N. Balakrishnan
Frequency Domain Robustness Analysis Of Hopfield And Modified Hopfield Neural Networks, Jie Shen, S. N. Balakrishnan
Mechanical and Aerospace Engineering Faculty Research & Creative Works
A variant of Hopfield neural network, called the modified Hopfield network, is formulated in this study. This class of networks consists of parallel recurrent networks which have variable dimensions that can be changed to fit the problem under consideration. It has a structure to implement an inverse transformation that is essential for embedding optimal control gain sequences. Equilibrium solutions of this network are discussed. The robustness of this network and the classical Hopfield network are carried out in the frequency domain using describing functions
Adaptive Critic Based Neural Networks For Control-Constrained Agile Missile Control, Dongchen Han, S. N. Balakrishnan
Adaptive Critic Based Neural Networks For Control-Constrained Agile Missile Control, Dongchen Han, S. N. Balakrishnan
Mechanical and Aerospace Engineering Faculty Research & Creative Works
We investigate the use of an `adaptive critic' based controller to steer an agile missile with a constraint on the angle of attack from various initial Mach numbers to a given final Mach number in minimum time while completely reversing its flightpath angle. We use neural networks with a two-network structure called `adaptive critic' to carry out the optimization process. This structure obtains an optimal controller through solving Hamiltonian equations. This approach needs no external training; each network along with the optimality equations generates the output for the other network. When the outputs are mutually consistent, the controller output is …
A Class Of Modified Hopfield Networks For Control Of Linear And Nonlinear Systems, Jie Shen, S. N. Balakrishnan
A Class Of Modified Hopfield Networks For Control Of Linear And Nonlinear Systems, Jie Shen, S. N. Balakrishnan
Mechanical and Aerospace Engineering Faculty Research & Creative Works
This paper presents a class of modified Hopfield neural networks (MHNN) and their use in solving linear and nonlinear control problems. This class of networks consists of parallel recurrent networks which have variable dimensions that can be changed to fit the problems under consideration. It has a structure to implement an inverse transformation that is essential for embedding optimal control gain sequences. Equilibrium solutions are discussed. Numerical results for a motivating aircraft control problem (linear) are presented. Furthermore, we formulate the state-dependent Riccati equation method (SDRE) for a class of nonlinear dynamical system and show how MHNN provides the solution. …
Adaptive Critic Based Neurocontroller For Autolanding Of Aircraft With Varying Glideslopes, Gaurav Saini, S. N. Balakrishnan
Adaptive Critic Based Neurocontroller For Autolanding Of Aircraft With Varying Glideslopes, Gaurav Saini, S. N. Balakrishnan
Mechanical and Aerospace Engineering Faculty Research & Creative Works
In this paper, adaptive critic based neural networks have been used to design a controller for a benchmark problem in aircraft autolanding. The adaptive critic control methodology comprises successive adaptations of two neural networks, namely `action' and `critic' networks until closed loop optimal control is achieved. The autolanding problem deals with longitudinal dynamics of an aircraft which is to be landed in a specified touchdown region in the presence of wind disturbances and gusts using elevator deflection as the control for glideslope and flare modes. The performance of the neurocontroller is compared to that of a conventional PID controller. Neurocontroller's …
A Dual Neural Network Architecture For Linear And Nonlinear Control Of Inverted Pendulum On A Cart, S. N. Balakrishnan, Victor Biega
A Dual Neural Network Architecture For Linear And Nonlinear Control Of Inverted Pendulum On A Cart, S. N. Balakrishnan, Victor Biega
Mechanical and Aerospace Engineering Faculty Research & Creative Works
The use of a self-contained dual neural network architecture for the solution of nonlinear optimal control problems is investigated in this study. The network structure solves the dynamic programming equations in stages and at the convergence, one network provides the optimal control and the second network provides a fault tolerance to the control system. We detail the steps in design and solve a linearized and a nonlinear, unstable, four-dimensional inverted pendulum on a cart problem. Numerical results are presented and compared with linearized optimal control. Unlike the previously published neural network solutions, this methodology does not need any external training, …
A New Neural Architecture For Homing Missile Guidance, S. N. Balakrishnan, Victor Biega
A New Neural Architecture For Homing Missile Guidance, S. N. Balakrishnan, Victor Biega
Mechanical and Aerospace Engineering Faculty Research & Creative Works
We present a new neural architecture which imbeds dynamic programming solutions to solve optimal target-intercept problems. They provide feedback guidance solutions, which are optimal with any initial conditions and time-to-go, for a 2D scenario. The method discussed in this study determines an optimal control law for a system by successively adapting two networks - an action and a critic network. This method determines the control law for an entire range of initial conditions; it simultaneously determines and adapts the neural networks to the optimal control policy for both linear and nonlinear systems. In addition, it is important to know that …
Use Of Hopfield Neural Networks In Optimal Guidance, S. N. Balakrishnan, James Edward Steck
Use Of Hopfield Neural Networks In Optimal Guidance, S. N. Balakrishnan, James Edward Steck
Mechanical and Aerospace Engineering Faculty Research & Creative Works
A Hopfield neural network architecture is developed to solve the optimal control problem for homing missile guidance. A linear quadratic optimal control problem is formulated in the form of an efficient parallel computing device known as a Hopfield neural network. Convergence of the Hopfield network is analyzed from a theoretical perspective, showing that the network, as a dynamical system approaches a unique fixed point which is the solution to the optimal control problem at any instant during the missile pursuit. Several target-intercept scenarios are provided to demonstrate the use of the recurrent feedback neural net formulation.
Approximate Analytical Guidance Schemes For Homing Missiles, S. N. Balakrishnan, Donald T. Stansbery
Approximate Analytical Guidance Schemes For Homing Missiles, S. N. Balakrishnan, Donald T. Stansbery
Mechanical and Aerospace Engineering Faculty Research & Creative Works
Closed form solutions for the guidance laws are developed using modern control techniques. The resulting two-point boundary value problem is solved through the use of the state transition matrix of the intercept dynamics. Results are presented in terms of a design parameter.
Decoupled Dynamics For Control And Estimation, S. N. Balakrishnan
Decoupled Dynamics For Control And Estimation, S. N. Balakrishnan
Mechanical and Aerospace Engineering Faculty Research & Creative Works
Decoupling of the dynamical equations in polar coordinates is used to develop a control scheme for use in target-intercept problems with passive measurements. By defining a pseudo control variable in the radial coordinate, the radial dynamics is made independent of the transverse dynamics. After solving for the radial control, the transverse control is determined through solutions to a two-point boundary value problem. Numerical results from a six degree-of-freedom simulation which used the decoupled control indicate that it is better than the completely Cartesian coordinate control for most of the cases considered. Decoupled control, though, is obtained iteratively through a two-point …