Artificial neuralnetworks offer an alternative strategy for the nonlinear control of unmanned underwater vehicles (UUVs). This paper investigates the use of a multi-layered perceptron (MLP) network in controlling an ...
详细信息
Artificial neuralnetworks offer an alternative strategy for the nonlinear control of unmanned underwater vehicles (UUVs). This paper investigates the use of a multi-layered perceptron (MLP) network in controlling an UUV over a sea-bed profile and compares the use of applying chemotaxis learning over that of the more commonly employed backpropagation algorithm. The results show for differing sized MLPs the chemotaxis algorithm produces a successful controller over the sea bed profile in an improved training time. To further vindicate the chemotaxis network, it was then presented with the problem of meeting a new profile to travel over. The results show from several simulation runs that the chemotaxis network provides a robust controller over numerous sea bed profiles of which it had no prior knowledge.< >
Describes the results of a simulation study of two fault detection approaches applied to a hydraulic test rig: an eigenstructure assignment approach; and a neural network based approach. The rig has a predominant natu...
详细信息
Describes the results of a simulation study of two fault detection approaches applied to a hydraulic test rig: an eigenstructure assignment approach; and a neural network based approach. The rig has a predominant natural frequency of the order of 20 Hz and is more representative of real industrial systems due to the presence of operating point nonlinearity, measurement noise and load disturbances.< >
Classifier systems lie midway between neuralnetworks and symbolic processing systems and potentially combine the benefits of both. They are parallel message-passing rule-based systems which use genetic algorithms to ...
详细信息
Classifier systems lie midway between neuralnetworks and symbolic processing systems and potentially combine the benefits of both. They are parallel message-passing rule-based systems which use genetic algorithms to discover new rules as well as providing for reinforcement learning and programming. It has been proposed that a suitable application of genetic algorithms is to evolve robots. A most suitable way to use genetic algorithms to evolve the controlsystems for robots is within the framework provided by classifier systems. At a SERC workshop on learning systems a number of groups presented successful applications of the genetic algorithm to control problems. However, one cannot evolve complex systems with a simple genetic algorithm nor is it wise or safe to start from scratch in real applications where programmed knowledge can provide constraints for the genetic algorithm to work within. If the genetic algorithm is to be used to evolve controlsystems for industrial or commercial applications one of the best ways to do this is within the framework of classifier systems.< >
Two distinctive approaches to the implementation of approximate reasoning in rule-based fuzzy decision-making and controlsystems are presented. While the first approach, based on possibility theory, bears some resemb...
详细信息
Two distinctive approaches to the implementation of approximate reasoning in rule-based fuzzy decision-making and controlsystems are presented. While the first approach, based on possibility theory, bears some resemblance to a traditional event-driven inference system with the incorporation of fuzzy concepts, the second scheme, based on the technique of neuralnetworks, represents a substantial departure from the traditional one and shows some promise in dealing with these fundamental issues. To demonstrate the applicability of the proposed approaches, a problem of multivariable fuzzy management of blood pressure has been studied using the simulation method.< >
Recently, there has been considerable interest in the use of artificial neuralnetworks for system identification and control. In this paper the authors discuss some constraints faced by using the Model-I and Model-II...
详细信息
Recently, there has been considerable interest in the use of artificial neuralnetworks for system identification and control. In this paper the authors discuss some constraints faced by using the Model-I and Model-II neural network systems introduced in work by K.S. Narendra and K. Parthasarathy (see ieeE Tranc. on neuralnetworks, vol.1, no.1, p.4-27 (1990)) for nonlinear system identification.< >
Artificial neuralnetworks attempt to model the massively parallel structure and the learning capability offered by biological neuralsystems. They have been shown to offer significant advantages over conventional pro...
详细信息
Artificial neuralnetworks attempt to model the massively parallel structure and the learning capability offered by biological neuralsystems. They have been shown to offer significant advantages over conventional processing techniques in certain recognition and control applications. However there are problems in selecting an appropriate neuron model and network configuration. The paper discusses these problems and identifies the need to establish a design methodology for neuralsystems.< >
Focuses on the use of adaptive connectionist networks for identification and control of nonlinear dynamic systems. control problems are usually solved using information about the controlled system and a series of proc...
详细信息
Focuses on the use of adaptive connectionist networks for identification and control of nonlinear dynamic systems. control problems are usually solved using information about the controlled system and a series of procedures to reach the desired control goal. From this point of view it is possible to decompose the control problem into two sub-problems: (1) a representation problem dealing with the way in which the information and dynamics of the system will be represented; and (2) a structural problem which concerns the definition of the architecture and the procedures which use the information on the system in order to reach the desired control goal. The connectionist approach has great potential for control applications since it addresses both problems, providing system representations and control procedures. The authors outline their work in both these areas.< >
The motivation for this study is derived from the potential problem of frost damage to citrus trees in the region surrounding Catania on the Island of Sicily Italy. The frost prediction technique developed during the ...
详细信息
The motivation for this study is derived from the potential problem of frost damage to citrus trees in the region surrounding Catania on the Island of Sicily Italy. The frost prediction technique developed during the investigation is based on the application of a parallel and distributed processing methodology to solve a time series weather forecasting problem. Past meteorological data observed and recorded in the region around Catania is used to train a neural network so that, given readings for a particular day, a prediction on whether frost formation is imminent can be made.< >
Addresses the use of a class of neural nets for the intelligent motion control and piloting of a variety of autonomous vehicles as part of an ESPRIT II mobile robotics project. Intelligent controllers are necessary in...
详细信息
Addresses the use of a class of neural nets for the intelligent motion control and piloting of a variety of autonomous vehicles as part of an ESPRIT II mobile robotics project. Intelligent controllers are necessary in order to cope with the vehicle complexities, internal parametric changes, safety imposed dynamic constraints as well as the effects of a dynamic environment. Single-layer, associative memory neuralnetworks, the modified Albus CMAC and B-splines, are proposed as the basis for an intelligent piloting system. These algorithms have an initially exponential convergence rate, are temporally stable (unlike the multilayer perceptron), noise resilient and exhibit known generalisation (interpolation) characteristics. Two alternative control architectures are presented and parallels are drawn with the more common fuzzy logic, radial basis functions and Kanerva's sparse distributed memory model.< >
Deals with the design of a neural network regulator (NNR) for nonlinear industrial dynamic systems, based on the multilayer perceptron (MLP) and using a hierarchically performed backpropagation (BP) algorithm. A novel...
详细信息
Deals with the design of a neural network regulator (NNR) for nonlinear industrial dynamic systems, based on the multilayer perceptron (MLP) and using a hierarchically performed backpropagation (BP) algorithm. A novel network architecture is employed for regulator design: the NNR consists of two subnetworks, one of which is used for I-O (input-output) mapping, while the other acts as an adaptive controller. The BP algorithm is employed to reproduce a nonlinear relation between the inputs and outputs of the plant and to update regulator parameters. The proposed architecture has the flexibility for adding more sensory information and facilitates extension to multi-input, multi-output systems and multivariable controllers. The operation of the NNR does not require a reference model or inverse system model, or any probing signals, and can produce more acceptable control signals than are obtained using the sign of the plant errors during the backpropagation procedure. The regulator has been applied to a complex nonlinear turbogenerator system.< >
暂无评论