{"id":"arxiv/2603.17632","url":"https://arxiv.org/abs/2603.17632","tags":["eess.SY","cs.RO","math.OC"],"score":0,"title":"Real-Time Online Learning for Model Predictive Control using a Spatio-Temporal Gaussian Process Approximation","domain":"eess.SY","source":"arxiv","authors":["Lars Bartels","Amon Lahr","Andrea Carron","Melanie N. Zeilinger"],"reviews":0,"subject":"eess.SY","summary":"Learning-based model predictive control (MPC) can enhance control performance by correcting for model inaccuracies, enabling more precise state trajectory predictions than traditional MPC. A common approach is to model unknown residual dynamics as a Gaussian process (GP), which leverages data and also provides an estimate of the associated uncertainty. However, the high computational cost of online learning poses a major challenge for real-time GP-MPC applications. This work presents an efficient implementation of an approximate spatio-temporal GP model, offering online learning at constant…","category":"research","comments":0,"disputed":false,"division":"eess","mentions":0,"discussed":0,"published":"2026-03-18","reactions":{}}