mlsanityData to 14 Oct 2026

Certifiable Safe Model-Based Reinforcement Learning with Control-Affine Dynamics Approximation

arXiv 2607.16501 ↗First posted 17 Jul 2026cs.ROcited by 0

On the map: Deep reinforcement learning methods → Safe and constrained RL

Abstract (arXiv)

Safe model-based reinforcement learning (RL) often bridges control-theoretic analysis and RL for robots to safely explore (partially) unknown system dynamics while deriving control actions for task efficiency. The control performance and safety assurance typically rely on prior knowledge of partially modeled nominal system dynamics and the data-driven models that compensate for residual model uncertainties. However, existing methods often overlook the structure of residual model uncertainties (e.g., components affine in control), which could lead to overly conservative robot behaviors or invalid safety guarantees under the safe learning-based controllers. This paper proposes a safe reinforcement learning framework that learns control-affine dynamics with a certifiable data-driven safe policy using control barrier functions (CBF). Specifically, we first use Control-Affine Random Fourier Features (ARFF) to model robot dynamics in a control-affine form, which offers computational efficiency that scales with dataset size and reduces potential model bias for model-based reinforcement learning. Then, a model-free, efficient uncertainty quantification method using adaptive conformal prediction (ACP) is applied to quantify the uncertainty in the safety constraint arising from the learned control-affine dynamics. This allows for data-driven safety assurance amenable to principled and efficient controller synthesis with CBF. Simulation results on the cartpole and the 3D quadrotor platforms demonstrate the effectiveness of the proposed framework.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Safe and constrained RL, one of 15 directions in Deep reinforcement learning methods. Builds on 22 of its 29 references that are in the map. First posted on arXiv 17 Jul 2026.

Builds on

22 of 29 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025Computationally and Sample Efficient Safe Reinforcement Learning Using Adaptive Conformal Prediction22 Mar

Other references · 21

  1. 2018Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Modelscited by 774
  2. 2016CVXPY: A Python-Embedded Modeling Language for Convex Optimizationcited by 641
  3. 2021Adaptive Conformal Inference Under Distribution Shiftcited by 375
  4. 2021Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learningcited by 371
  5. 2017Safe Model-based Reinforcement Learning with Stability Guaranteescited by 366
  6. 2019End-to-End Safe Reinforcement Learning through Barrier Functions for Safety-Critical Continuous Control Taskscited by 289
  7. 2022Conformal Inference for Online Prediction with Arbitrary Distribution Shiftscited by 128
  8. 2022Safe Planning in Dynamic Environments using Conformal Predictioncited by 119
  9. 2017Information Theoretic Model Predictive Control: Theory and Applications to Autonomous Drivingcited by 90
  10. 2020Information Theoretic Regret Bounds for Online Nonlinear Controlcited by 90
  11. 2020Efficient Model-Based Reinforcement Learning through Optimistic Policy Search and Planningcited by 86
  12. 2021Safe Reinforcement Learning Using Robust Control Barrier Functionscited by 59
9 more
  1. 2017Data-Efficient Reinforcement Learning with Probabilistic Model Predictive Controlcited by 53
  2. 2022Adaptive Conformal Prediction for Motion Planning among Dynamic Agentscited by 53
  3. 2017Safe Learning of Quadrotor Dynamics Using Barrier Certificatescited by 35
  4. 2019On Online Learning in Kernelized Markov Decision Processescited by 35
  5. 2021Bias-Free Scalable Gaussian Processes via Randomized Truncationscited by 16
  6. 2024Safety-Critical Control with Uncertainty Quantification using Adaptive Conformal Predictioncited by 7
  7. 2024Random Features Approximation for Control-Affine Systemscited by 4
  8. 2022Sample-efficient Safe Learning for Online Nonlinear Control with Control Barrier Functionscited by 3
  9. 2023Trigonometric Quadrature Fourier Features for Scalable Gaussian Process Regressioncited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026CAPSULE: Control-Theoretic Action Perturbations for Safe Uncertainty-Aware Reinforcement Learning
  2. 2026Sampling-Based Safe Reinforcement Learning
  3. 2025Multi-Constraint Safe Reinforcement Learning via Closed-form Solution for Log-Sum-Exp Approximation of Control Barrier Functions
  4. 2025Safe model-based Reinforcement Learning via Model Predictive Control and Control Barrier Functions
  5. 2021Safe Continuous Control with Constrained Model-Based Policy Optimization
  6. 2024Handling Long-Term Safety and Uncertainty in Safe Reinforcement Learning
  7. 2026Robust Koopman Control Barrier Filters for Safe Actor-Critic Reinforcement Learning
  8. 2024Disturbance Observer-based Control Barrier Functions with Residual Model Learning for Safe Reinforcement Learning