{"ID":2831848,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.08013","arxiv_id":"2512.08013","title":"Learning Dynamics from Infrequent Output Measurements for Uncertainty-Aware Optimal Control","abstract":"Reliable optimal control is challenging when the dynamics of a nonlinear system are unknown and only infrequent, noisy output measurements are available. This work addresses this setting of limited sensing by formulating a Bayesian prior over the continuous-time dynamics and latent state trajectory in state-space form and updating it through a targeted Metropolis-Hastings sampler equipped with a numerical ODE integrator. The resulting posterior samples are used to formulate a scenario-based optimal control problem that accounts for the uncertainty in the dynamics and latent state and is solved using standard nonlinear programming methods. The approach is validated in a numerical case study on glucose regulation using a Type 1 diabetes model.","short_abstract":"Reliable optimal control is challenging when the dynamics of a nonlinear system are unknown and only infrequent, noisy output measurements are available. This work addresses this setting of limited sensing by formulating a Bayesian prior over the continuous-time dynamics and latent state trajectory in state-space form...","url_abs":"https://arxiv.org/abs/2512.08013","url_pdf":"https://arxiv.org/pdf/2512.08013v2","authors":"[\"Robert Lefringhausen\",\"Theodor Springer\",\"Sandra Hirche\"]","published":"2025-12-08T20:10:37Z","proceeding":"eess.SY","tasks":"[\"eess.SY\",\"cs.LG\",\"math.OC\"]","methods":"[]","has_code":false}