In engineering application, the characteristics of the control system are entirely determined by the system controller once the controlled object has been chosen. Improving the traditional controller or constructing the new controller is an unfading study field of control theory and application. The control system is greatly enriched and developed by this way. As a complicated self-adaptable system, the biological immune system can effectively and smoothly stand against antigens and viruses intruded into organism. It is possible to improve the self-learning, adaptive and robustness capability of the control system through embedded an artificial immune controller in control system. Based on the biological immune mechanism and artificial immune model, this chapter attempts to study the immune controller design and application in traditional control system..First, a kind of artificial immune controller is proposed based on the T-B cells immunity. The boundedness and the stability of SISO control systems, which constructed by the artificial immune controller, are proved by the little gain theorem. A general controller structure frame based on the T-B cells immunity is proposed, which includes the same kind of controller proposed previously. The validity of this artificial immune controller is verified by simulation. Second, a new type of artificial immune controllers is constructed according to a simple double-cell immune dynamics model. The non-error characteristic of SISO control systems, which constructed by the artificial immune controller, is proved by the nonlinear theory in this chapter. The I/O stability and no-error characteristic of the system are verified by simulations, which show that the kind of artificial immune system have good anti-lag capability. Third, the Varela immune network model has been improved based on which an artificial immune system is proposed. The odd linearization method of the non-linear system is used to prove the stability and non-error characteristic of the SISO system constructed by the artificial immune control system. Its I/O stability, non-error characteristic and strong anti-lag capability are also verified by simulation. Finally, based on the comparison of the three kinds of immune controllers, a general structure of the artificial immune controller is proposed. The further study on this field is indicated in this chapter lastly.
In real applications, once the controlled object and the measuring components have been determined, the performance of linear control systems will depend on controllers. Therefore, designing and constructing new types of controllers or improving traditional controllers have been being a fascinating topic in the theory of control systems. For instance, one of the important applications of fuzzy control theory is to design and construct controllers which have great adaptability and robustness(Jingzhen & Zengji, 1997), so is the neural network (Dongqiang & Yaqing, 2001) and expert system. Even the widely used PID controllers are also studied by many individuals to make it more intelligent. The study and application of intelligent controllers has extraordinarily enriched and developed the theory of control systems.
Intelligent controllers fall into four levels according to the designing complexity: simple robust feedback control, parameter adaptive control based on error criterion, adaptive control based on optimization of objective function and adaptive control based on global optimization of objective function varied with environment conditions. Corresponding with these, biologic immune system fall into four similar levels: firstly, the innate immune system is corresponding to robust feedback control; secondly, T cells stimulate B cells to resist antigens; thirdly, microphages present antigens to T cells and activate B cells; lastly, microphages have some changes against certain antigens. Studies have found that biologic immune system has a series of excellent control characteristics, such as adaptability, distributed control and coordination, dynamic, robust, self-organization, self-learning etc. So a further study of immune system on its mechanism and strategies will provide some ideas for constructing advanced intelligent controllers.
Key Terms in this Chapter
Artificial Immune Models: Artificial Immune models are a kind of mathematical models which are constructed by macro-biological or micro-biological immune mechanism.
Antigen: The antigen is a substance which stimulate the person or animal body to produce sensitized antibodies or lymphocytes, these products arose specific responses in the person or animal body. Antigens include toxins, bacteria, foreign blood cells, the cells of transplanted organs and so on.
Immune Factor: The immune factors are a sort of small molecule polypeptide which mainly were secreted the biological activity immunocyte, such as transfer factor, distinguishing factor and so on. Immune factors mainly make regulative roles in the immune recognition and immune responses.
Artificial Immune Control: Artificial immune control is similar ordinary control except Artificial immune algorithm and model in it. Artificial immune control is the process to detect, control or manage the applications, such as controlled objects, production procedures, or managements, by constructing immune models or algorithms suitable for control systems through reference of measures and regulars extracted from immune mechanisms, models and the cooperation between immune factors.
Immune Controller: Immune controller is a man-made controller that based on measures and regulars of macro immune mechanisms, models or micro cooperation between immune factors to improve the performance of a control system. Immune controllers mostly are nonlinear dynamic models or algorithms with iteration process. Artificial immune controller is similar ordinary controller except immune algorithm and model in it.
Artificial Immunity: The algorithms or dynamics models which are founded by men imitate the part or all of the biological immune mechanism to gain some special features or peculiar functions. Those algorithms or the dynamics models are usually known as the artificial immunity.
Antibody: Antibodies are protein substances produced in the blood or tissues to response to a specific antigen, such as a bacterium or a toxin. Antibodies destroy or weaken bacteria and neutralize organic poisons, thus forming the basis of immunity.