This paper describes a learning augmented recursive estimation approach for nonlinear dynamical systems having unmodeled nonlinearities. Utilizing a passive spatially-localized learning system, an approximation of the...
详细信息
This paper describes a learning augmented recursive estimation approach for nonlinear dynamical systems having unmodeled nonlinearities. Utilizing a passive spatially-localized learning system, an approximation of the unknown nonlinearity is synthesized online, based on state and parameter estimates from a nonlinear recursive estimator (an adaptive form of the extended Kalman filter). The learned model of the nonlinearity is used, in turn, to improve the performance of the recursive estimator. We demonstrate the approach on a second-order, mass-spring-damper system, where the spring stiffness is a nonlinear function of position. Simulation results indicate that, relative to more traditional adaptive estimation schemes, markedly improved estimation performance can be achieved.
This paper describes two computing paradigms known as neural computing and evolutionary computing, and their potential contribution to building intelligent software systems. The paper begins by giving a brief introduc...
详细信息
This paper describes two computing paradigms known as neural computing and evolutionary computing, and their potential contribution to building intelligent software systems. The paper begins by giving a brief introduction to the origins of each paradigm. Then two sections introduce the basic principles, and identify the role of each paradigm in intelligent system design. Each section ends with a number of applications that have been or are being investigated. These include connection admission control, modem communication, adaptive model-based control, face and handwriting recognition, frequency assignment, help-desk scheduling, financial time-series prediction, face recognition and evolving agent behavior. A section introduces the idea of using communications theory to design neural networks and the paper concludes with the authors' views on the future of neural and evolutionary computing for intelligent software systems.
This paper proposes a method of 3-dimensional measurement of a planar surface by using two fixed light sources and a TV camera. A set of two images is recorded by the camera switching on each light alternately. The pe...
详细信息
This paper proposes a method of 3-dimensional measurement of a planar surface by using two fixed light sources and a TV camera. A set of two images is recorded by the camera switching on each light alternately. The peak of the luminance distribution in each image is detected, and the orientation and distance of the planar surface are calculated. The position of the peak of a luminance distribution can be estimated accurately by using an image processing. The light sources can be conventional apparatus with no particular structure. The method is simple and suitable for a vision system on an indoor mobile robot.
Visibility analysis algorithms use digital elevation models (DEMs), which represent terrain topography, to determine visibility at each point on the terrain from a given location in space. This analysis can be computa...
Visibility analysis algorithms use digital elevation models (DEMs), which represent terrain topography, to determine visibility at each point on the terrain from a given location in space. This analysis can be computationally very demanding, particularly when manipulating high resolution DEMs accurately at interactive response rates. Massively data-parallel computers offer high computing capabilities and are very well-suited to handling and processing large regular spatial data structures. In the paper, the authors present a new scanline-based data-parallel algorithm for visibility analysis. Results from an implementation onto a MasPar massively data-parallel SIMD computer are also presented.
Learning of large-scale neural networks suffers from computational cost and the local minima problem. One solution to these difficulties is the use of modular structured networks. Proposed here is the learning of modu...
详细信息
Learning of large-scale neural networks suffers from computational cost and the local minima problem. One solution to these difficulties is the use of modular structured networks. Proposed here is the learning of modular networks using structural learning with forgetting. It enables the formation of modules. It also enables automatic utilization of appropriate modules from among the previously learned ones. This not only achieves efficient learning, but also makes the resulting network understandable due to its modular character. In the learning of a Boolean function, the present module acquires information from its subtask module without any supervision. In the parity problem, a previously learned lower-order parity problem is automatically used. The geometrical transformation of figures can be realized by a sequence of elementary transformations. This sequence can also be discovered by the learning of multi-layer modular networks. These examples well demonstrate the effectiveness of modular structured networks constructed by structural learning with forgetting.
Many physical systems that occur in applications are naturally passive, for example, mechanical systems with dual sensors and actuators, and electrical circuits with passive components. Taking advantage of this proper...
详细信息
Many physical systems that occur in applications are naturally passive, for example, mechanical systems with dual sensors and actuators, and electrical circuits with passive components. Taking advantage of this property, many controller schemes have been proposed with the property that the controller is strictly positive real. Due to design and implementation considerations, the plant or the controller may need to be approximated by a lower-order system. It is highly desirable for the reduced-order system to also possess the positive realness property to guarantee that the resulting closed-loop system remains stable. Motivated by this problem, this paper considers the general model-reduction problem for a positive real system under the constraint that the reduced system is also positive real. We present a solution based on the balanced stochastic truncation. When the higher-order system is strictly positive real, we derive an H(infinity) norm bound on the approximation error. We also consider alternate approaches of approximating the spectral factors with associated H(infinity) norm error bounds. An example is included to show the efficacy of this method and comparison with other approaches.
A series of piecewise multiple general orthogonal polynomials (PMGOPs) is introduced and applied to the parameter identication problem for a class of continuous nonlinear systems. After introducing PMGOPs and discussi...
详细信息
A series of piecewise multiple general orthogonal polynomials (PMGOPs) is introduced and applied to the parameter identication problem for a class of continuous nonlinear systems. After introducing PMGOPs and discussing their main properties, an effective procedure for the parameter identification of a large class of continuous systems, called parameter separable systems, is proposed. Some relevant algorithms are presented. The procedure given in the paper has the following advantages compared with other methods: the identification algorithm (IA) obtained is computationally fast and accurate;the IA can be implemented in a recursive fashion;the IA is effective for a small number of data points;the IA does not require a priori knowledge of the estimated parameters;the IA is tolerant to choices of expanding orders when PMGOPs are applied. Some numerical examples, together with an experimental example taken from a biochemical fermentation process, are given for illustrative purposes.
Reactive ion etching is an important process in the fabrication of microelectronic devices. This article reports on work in progress towards developing a strategy for controlling sidewall profile during this process. ...
详细信息
Reactive ion etching is an important process in the fabrication of microelectronic devices. This article reports on work in progress towards developing a strategy for controlling sidewall profile during this process. In this strategy, a response surf...
The task of guaranteeing that processes meet their deadlines is one of the most complex in the design of hard real-time controlsystems. This paper describes how process scheduling has been integrated into a framework...
详细信息
The task of guaranteeing that processes meet their deadlines is one of the most complex in the design of hard real-time controlsystems. This paper describes how process scheduling has been integrated into a framework of tools for the development of such systems. Functional specifications in the form of control system block diagrams are translated by the Framework into real-time source code which may be executed in a parallel processing environment. The automated off-line scheduling tool takes process specifications from the system under development and attempts to map these processes to the processors in such a way that all the system deadlines are met by using a detailed model of the run-time environment.
The robust stability and performance of a binary distillation column model is analysed using a minimal state space representation. The uncertainty modelling of the column is resulted in an LFT form with a four element...
详细信息
The robust stability and performance of a binary distillation column model is analysed using a minimal state space representation. The uncertainty modelling of the column is resulted in an LFT form with a four element diagonal block structure taking into account the dependence between the steady state gains. Robust analysis of the distillation column dynamics is performed and compared using µ - and l 1 -analysis tools. The magnitude of possible overshoots is determined by l 1 -test for robust performance and the dependence of this magnitude on the dominant time constant (i.e. the purity of the separation) is investigated. Futhermore the effect of the mi certainties in the time constants and the singular values on the robust performance is also analysed using l 1 -test. It has been found that the uncertainty in the time constants affects dominantly the robust performance of the open-loop system, i.e. the magnitude of possible overshoots.
暂无评论