What is the difference between direct and indirect learning control? I found the following comments on direct and indirect control in this paper by Wang, Gao, and Doyle: "Survey on iterative learning control, repetitive control and run-to-run control." My understanding is that in indirect control you can alter the control parameters and the input signal as opposed to just the input signal in direct control; is this correct? I also do not understand the significance of a "local" controller.
There are two application modes to use the learning-type control. First, the learning-type control method is used to determine the control signal directly, and this kind of learning-type control is called direct learning-type control. Second, there is a local feedback controller in each cycle and the learning-type control is used to update the parameter settings of the local controller, so this kind is called indirect learning-type control. The methods that can be used for designing direct learning-type control and indirect learning-type control will be discussed in Section 4 and in Section 5, respectively.
•Wang, Gao, Doyle, Survey on iterative learning control, repetitive control and run-to-run control
In adaptive control, direct control is the case where the controller parameters are adjusted according to some criterion in order to achieve a control objetive. In indirect control, the parameters of a model of the process to be controlled are adjusted, with the goal of obtaining a good match between observed process data and the model; the model is in turn used to design a controller, just as you would do offline if you knew that the model is a good one.
I have not read the paper you refer to, but I suppose that the usage is similar.