Loading...
Please wait, while we are loading the content...
Similar Documents
A new approach for designing self-organizing systems and application to adaptive control
| Content Provider | Semantic Scholar |
|---|---|
| Author | Ramamoorthy, P. A. Zhang, Shi Lin, Yueqing Huang, Song |
| Copyright Year | 1993 |
| Abstract | There is tremendous interest in the design of intelligent machines capable of autonomous learning and skillful performance under complex environments. A major task in designing such systems is to make the system plastic, and adaptive when presented with new and useful information and stable in response to irrelevant events. A great body of knowledge, based on neuro-physiological concepts, has evolved as a possible solution to this problem. Adaptive resonance theory (ART) is a classical example under this category. The system dynamics of an ART network is described by a set of differential equations with nonlinear functions. An entirely new approach for designing self-organizing networks characterized by nonlinear differential equations is proposed in this paper. Similar to the neuro-physiologicai approach, the method presented here relies upon another area that of passive nonlinear network theory. A passive nonlinear network is formed by proper interconnection of various nonlinear elements where each and every nonlinear element is constrained to be lossless or lossy. When energy storing elements are present in such a network, we can obtain a set of Input/Output relationships as nonlinear differential equations. The basic property that the network is lossy (consumes energy) ensures that the nonlinear differential equations obtained from the network would represent absolutely stable systems and this property holds as long as the individual element values are maintained in their permissible range of values. Thus, to deign complex nonlinear systems (a complex nonlinear plant plus a controller to optimize its performance, for example) and self-organizing systems, one simply has to force the system dynamics to mimic the dynamics of a properly constructed passive nonlinear network, a process akin to reverse engineering. In our research which is in its early stages, we have developed the basis for the above approach and applied it with relative ease to a number of problems leading to encouraging results. The fruits of such an approach seem to be endless. For example, the approach can be applied to linear and nonlinear controller design (for linear and nonlinear plants), self tuning controllers, model reference adaptive controllers, self-organizing networks, adaptive IIR filter design, adaptive beam-forming, two-dimensional systems, fuzzy systems etc. In this paper, we provide some details of this approach and show results from some of these topics to show the power of this approach. 295 https://ntrs.nasa.gov/search.jsp?R=19930013026 2018-05-12T07:38:53+00:00Z |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19930013026.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
