{"id":"arxiv/2606.17766","url":"https://arxiv.org/abs/2606.17766","tags":["math.OC"],"score":0,"title":"Operator Learning for PDE Backstepping Control of Parabolic Equations on Time-Varying Domains","domain":"math.OC","source":"arxiv","authors":["Jinrun Yan","Kai Liu","Zhong-Jie Han"],"reviews":0,"subject":"math.OC","summary":"This paper develops a learning-based boundary control framework for stabilizing a parabolic equation defined on time-varying spatial domain. Although the partial differential equation (PDE) backstepping method provides a systematic theoretical framework for such moving-boundary systems, its real-time implementation is hindered by the need to repeatedly solve time-varying kernel PDEs on evolving domains. To overcome this limitation, we first formulate the time-varying backstepping design as an operator that maps the moving-boundary trajectory to the corresponding backstepping kernel. By…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-06-16","reactions":{}}