Abstract

Iterative learning control is applicable to systems that repeat the same finite-duration task over and over again. The first research in this area is widely credited to Arimoto et al in 1984. The core idea is that once the execution of the task is complete, all information generated over its duration is available to update the control signal for the next execution. The original work was in robotics, where the pick and place operation satisfied the ILC ’s definition. Since then, many other areas of application have arisen, including healthcare. One notable feature is that in many cases model based design has been followed through to at least laboratory based experimental validation. This seminar will have this theme, i.e., control theory plus results from at least laboratory validation. The speakers thoughts on the future of this area will also be covered, including the `links’ with reinforcement learning.

The seminar will be held in the LR3A , Department of Engineering, and online (zoom): https://newnham.zoom.us/j/92544958528?pwd=YS9PcGRnbXBOcStBdStNb3E0SHN1UT09