抄録
This short note gives a rigorous condition and proof for convergence of learning control of linear time-invariant systems, especially for the case of not using the derivative of the error signal. The proof also clarifies the existing confusion in the literature.
