![]() |
|||||||
|
|||||||
![]() |
![]() |
|
Introduction to Adaptive
Inverse Control: The principal advantage of adaptive inverse control is its combination of simplicity and power. Adaptive inverse control has been shown to be optimal under the conventional GLS assumptions. But more importantly, it can be easily extended to cases that go beyond these assumptions, because of the natural way with which neural networks can be incorporated in the algorithms. The structure of the calculations in adaptive inverse control allow us to create a development environment that can be used on the factory floor using a combination of an embedded on-line microcontroller with an off-line parameter optimization algorithm. Most control algorithms are based on the feedback mechanism. The idea of adaptive inverse control is different from this. The controller is in series with the plant, but there is no direct feedback from the plant output back to the control input. However, the method is NOT strictly feedforward (open-loop) control since the controller parameters are being adapted at ALL times with information from the output of the plant and the command input. This leads to very interesting deployment solutions as we will demonstrate next. From a practical perspective, the key advantages of inverse adaptive control are:
The basic inverse adaptive control technique is shown in Figure 1. The top block diagram shows the training of the inverse model. Essentially, the neural network is learning to recreate the input that created the current output of the plant (ventilator/patient combination in our case). Once properly trained, the inverse model can be used to control the plant since it can create the necessary control signals to create the desired system output.
Figure 1 - Training and Use of an Inverse Controller |
|
Product questions? Contact info@nd.com Website questions? Contact webmaster@nd.com |