Challenge II

Introduction

Recently, a growing number of tasks for which is appropriate or necessary to use autonomous adaptive control system. However, under the circumstances, the constant tendency to increase the complexity and variety of control tasks, the existing approaches to the development of adaptive control systems are no longer able to provide the necessary level of control and adaptation.

Self-guided, or adaptive, system is, in fact, the hardware and software computer systems, operating independently from the man. It is their fear some fans staging of "philosophical" questions that draw in the imagination computer, released from the authority. We may recall the classic science fiction and marvel at the accuracy of the time and predictions of the researchers. The control system industry begins to actively discuss the concept of adaptive systems, to develop now the leading autonomous systems.

In this paper, we are going to implement an adaptive control system using Simulink for a step input passing through a transfer function. Also, the response of the system will be further improved by using the suitable techniques to reduce the offset error.

Discussion

Analysis of the current state of affairs in the field of autonomous adaptive control systems reveals the following main problems faced by the developers of these systems:

The problem of coordinating the various forms of behavior and organization of interaction between them.

The problem of planning actions. The control system must not only respond to the current situation, but also to predict the sequence of actions that will lead to the desired goal.

The problem of learning from the experience of their interaction with the environment. This problem involves not only the form of a separate study of behavior, but the issues of training interaction between different forms of behavior.

The problem of the formation of new types of behavior. The control system must be able to independently create new behaviors to achieve new goals, and incorporate them into the overall structure of control.

The problem of universality. Systems of various objects should follow the same principles of management and be based on a single model (brown,edu, 2013).

Existing approaches currently building management systems only solve some of these problems. Any general and sufficiently universal models that would solve all of these problems at the present time and is not offered.

Right now, most researchers agree that the creation of adaptive control system is advantageously carried out, starting from the general theories and conceptual schemes. One of these concepts is a general theory of functional systems. Despite the fact that this theory was originally introduced to describe the general scheme of controlling the behavior of living organisms based on it is possible to offer a total cybernetic model of adaptive management purposeful behavior.

In this paper, we propose a model adaptive control system based on the formalization of the basic ideas of the theory of functional systems using feedback methods to extract knowledge for learning from the experience of their interaction with the environment. The proposed model has been successfully used to create an ...