Answers are grounded in this paper and cite their sources.
In this paper, through the analysis on error contraction conditions in the frequency domain, it is found that a negative learning gain can be used in iterative learning control (ILC). The proper use of negative learning gain can increase the learnable frequency range. The conventional P-type learning control with both positive and negative gains is used over multichannel learnable bands. The design procedure is demonstrated via an example of robot joint control. Two channels, one uses a positive learning gain and the other a negative learning gain, substantially increase the frequency components to be learned by the P-type controller. Experimental results verify the effectiveness of using a negative learning gain.
No takes yet. Share an insight, caveat, or question.
Ye et al. (2006) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: