mlsanityData to 14 Oct 2026

CoVO-MPC: Theoretical Analysis of Sampling-based MPC and Optimal Covariance Design

arXiv 2401.07369 ↗First posted 14 Jan 2024cs.LG cs.ROcited by 4

On the map: Deep reinforcement learning methods → RL for autonomous driving and traffic

Abstract (arXiv)

Sampling-based Model Predictive Control (MPC) has been a practical and effective approach in many domains, notably model-based reinforcement learning, thanks to its flexibility and parallelizability. Despite its appealing empirical performance, the theoretical understanding, particularly in terms of convergence analysis and hyperparameter tuning, remains absent. In this paper, we characterize the convergence property of a widely used sampling-based MPC method, Model Predictive Path Integral Control (MPPI). We show that MPPI enjoys at least linear convergence rates when the optimization is quadratic, which covers time-varying LQR systems. We then extend to more general nonlinear systems. Our theoretical analysis directly leads to a novel sampling-based MPC algorithm, CoVariance-Optimal MPC (CoVo-MPC) that optimally schedules the sampling covariance to optimize the convergence rate. Empirically, CoVo-MPC significantly outperforms standard MPPI by 43-54% in both simulations and real-world quadrotor agile control tasks. Videos and Appendices are available at urlhttps://lecar-lab.github.io/CoVO-MPC/.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in RL for autonomous driving and traffic, one of 15 directions in Deep reinforcement learning methods. Builds on 24 of its 46 references that are in the map. First posted on arXiv 14 Jan 2024.

Builds on

24 of 46 references.

References · 24

  1. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  2. 2018Deep Reinforcement Learning in a Handful of Trials using Probabilistic Dynamics Modelscited by 774
  3. 2019When to Trust Your Model: Model-Based Policy Optimizationcited by 773
  4. 2019Model-Based Reinforcement Learning for Ataricited by 502
  5. 2022Temporal Difference Learning for Model Predictive Controlcited by 397
  6. 2018Visual Foresight: Model-Based Deep Reinforcement Learning for Vision-Based Robotic Controlcited by 230
  7. 2020Model-Based Offline Planningcited by 129
  8. 2018Plan Online, Learn Offline: Efficient Learning and Exploration via Model-Based Controlcited by 113
  9. 2023Reaching the Limit in Autonomous Racing: Optimal Control versus Reinforcement Learningcited by 111
  10. 2018Neural Lander: Stable Drone Landing Control using Learned Dynamicscited by 85
  11. 2022Neural-Fly Enables Rapid Learning for Agile Flight in Strong Windscited by 85
  12. 2021Temporal Predictive Coding For Model-Based Planning In Latent Spacecited by 55
12 more
  1. 2020The Power of Predictions in Online Controlcited by 31
  2. 2019An Online Learning Approach to Model Predictive Controlcited by 30
  3. 2021Perturbation-based Regret Analysis of Predictive Control in Linear Time Varying Systemscited by 25
  4. 2021Regularized Newton Method with Global O(1/k2) Convergencecited by 22
  5. 2020Neural-Swarm2: Planning and Control of Heterogeneous Multirotor Swarms using Learned Interactionscited by 20
  6. 2019Information Theoretic Model Predictive Q-Learningcited by 18
  7. 2021Robust Model Predictive Path Integral Control: Analysis and Performance Guaranteescited by 17
  8. 2023DATT: Deep Adaptive Trajectory Tracking for Quadrotor Controlcited by 16
  9. 2020L1-Adaptive MPPI Architecture for Robust and Agile Control of Multirotorscited by 14
  10. 2021Variational Inference MPC using Tsallis Divergencecited by 11
  11. 2018Vision-Based High Speed Driving with a Deep Dynamic Observercited by 7
  12. 2021Constrained Covariance Steering Based Tube-MPPIcited by 2

Built on by

Cited by year

Citing papers per year: 2024: 1; 2025: 3.

4 papers.

  1. 2024Model-Based Diffusion for Trajectory Optimizationcited by 32
  2. 2025Optimality and Suboptimality of MPPI Control in Stochastic and Deterministic Settingscited by 1
  3. 2025Model Tensor Planningcited by 0
  4. 2025PegasusFlow: Parallel Rolling-Denoising Score Sampling for Robot Diffusion Planner Flow Matchingcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022Learning to Optimize in Model Predictive Control
  2. 2025Gauss-Newton accelerated MPPI Control ↗
  3. 2024Faster Model Predictive Control via Self-Supervised Initialization Learning
  4. 2026A Systematic Review and Taxonomy of Reinforcement Learning-Model Predictive Control Integration for Linear Systems
  5. 2025Unifying Model Predictive Path Integral Control, Reinforcement Learning, and Diffusion Models for Optimal Control and Planning
  6. 2024Equivariant Action Sampling for Reinforcement Learning and Planning
  7. 2024Towards an Adaptable and Generalizable Optimization Engine in Decision and Control: A Meta Reinforcement Learning Approach
  8. 2025Synthesis of Model Predictive Control and Reinforcement Learning: Survey and Classification