Frontend
GP--RL--MPC: Uncertainty-Aware Reinforcement Learning-Augmented Model Predictive Control with Gaussian Process Postdecision State-Value Functions
arXiv eess.SY0 views
arXiv:2611.22038v1 Announce Type: cross Abstract: We propose Gaussian Process Reinforcement Learning-Augmented Model Predictive Control (GP--RL--MPC), which uses a learned critic as the MPC terminal penalty to account for costs beyond the prediction horizon. Gaussian process regression (GPR) enables data-efficient critic learning from limited, noisy rollouts within optimistic approximate policy iteration (OAPI), with soft updates mitigating policy chattering. A tunable risk coefficient weights GP epistemic uncertainty in the terminal cost, regulating sensitivity to uncertain critic predictions. The postdecision state formulation shifts disturbance averaging offline to account for aleatoric uncertainty from the terminal transition onward. Under matching objectives and admissible actions, a one-step horizon recovers greedy value-space RL. After fitting GP hyperparameters by exact marginal likelihood, we construct a sparse deterministic training conditional (DTC) posterior using inducing points selected by pivoted Cholesky (PC), reducing online evaluation costs as the training dataset grows. In a continuous stirred-tank reactor (CSTR) subject to exogenous disturbances, GP--RL--MPC with an appropriately selected risk coefficient and policy screening reduces the sum of squared errors (SSE) by an average of 80.52% relative to an otherwise matched baseline MPC across five random seeds (sample standard deviation: 3.48%).
Source:https://arxiv.org/abs/2611.22038PreviousCausal Input--Output Modeling with Prior InformationNextCRA: Continuous Risk Assessment for Safety-critical Visual Navigation
Related posts
How I Built 363 Free Web Calculators with Next.js & Tailwind (Zero Ads, Instant Math)
0dev.to
Our team name generator draws 6 of 110 names uniformly, and you will still see a repeat by the third press
0dev.to
HotSymbol: A Simple Way to Find, Copy, and Use Thousands of Symbols Online
0dev.to
No Webfonts, No Frameworks: The Performance Budget I Keep on My Tool Pages
0dev.to
How the Turkish ID number (TCKN) checksum works, with validators in JavaScript, Python and PHP
0dev.to
Our homepage demo has four numbers for one score, and an ordinal suffix with three branches that can never run
0dev.to