Iterative learning control is an adaptive, feedforward control technique traditionally used to improve the performance of systems that execute a task repetitively. While generally applied to systems driven by temporal dynamics, there exist applications, such as additive manufacturing, for which spatial dynamics play a particularly important role in determining system behavior. To ensure high fidelity functionality for these application spaces, this paper presents a spatial learning framework for optimizing multiple performance metrics simultaneously. Utilizing a one-step optimization approach enables direct evaluation of design trade-offs over a broad range of potential solutions. The multi-objective spatial learning framework, along with stability and convergence analysis is presented. Simulation results validate the control framework.

This content is only available via PDF.
You do not currently have access to this content.