mlsanityData to 7 Oct 2026

Model Predictive Adversarial Imitation Learning for Planning from Observation

arXiv 2507.21533 ↗First posted 29 Jul 2025cs.RO cs.AIcited by 2

On the map: Offline RL and imitation learning from data → Imitation learning methods

Abstract (arXiv)

Human demonstration data is often ambiguous and incomplete, motivating imitation learning approaches that also exhibit reliable planning behavior. A common paradigm to perform planning-from-demonstration involves learning a reward function via Inverse Reinforcement Learning (IRL) then deploying this reward via Model Predictive Control (MPC). Towards unifying these methods, we derive a replacement of the policy in IRL with a planning-based agent. With connections to Adversarial Imitation Learning, this formulation enables end-to-end interactive learning of planners from observation-only demonstrations. In addition to benefits in interpretability, complexity, and safety, we study and observe significant improvements on sample efficiency, out-of-distribution generalization, and robustness. The study includes evaluations in both simulated control benchmarks and real-world navigation experiments using few-to-single observation-only demonstrations.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Imitation learning methods, one of 13 directions in Offline RL and imitation learning from data. Builds on 51 of its 67 references that are in the map. First posted on arXiv 29 Jul 2025.

Builds on

51 of 67 references.

Closest ancestors · 9

Same direction, newest first

  1. 2024Reward-free World Models for Online Imitation Learning17 Oct
  2. 2024C-GAIL: Stabilizing Generative Adversarial Imitation Learning with Control Theory26 Feb
  3. 2022Planning for Sample Efficient Imitation Learning18 Oct
  4. 2022Goal-Aware Generative Adversarial Imitation Learning from Imperfect Demonstration for Robotic Cloth Manipulation21 Sep
  5. 2021What Matters for Adversarial Imitation Learning?1 Jun
  6. 2021Hyperparameter Selection for Imitation Learning25 May
  7. 2019A Divergence Minimization Perspective on Imitation Learning Methods6 Nov
  8. 2019Recent Advances in Imitation Learning from Observation30 May
  9. 2018Discriminator-Actor-Critic: Addressing Sample Inefficiency and Reward Bias in Adversarial Imitation Learning9 Sep

Other references · 42

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2015Continuous control with deep reinforcement learningcited by 3,279
  3. 2015Trust Region Policy Optimizationcited by 2,503
  4. 2015High-Dimensional Continuous Control Using Generalized Advantage Estimationcited by 1,916
  5. 2018Spectral Normalization for Generative Adversarial Networkscited by 1,618
  6. 2016Generative Adversarial Imitation Learningcited by 1,576
  7. 2020Energy-based Out-of-distribution Detectioncited by 1,296
  8. 2017Learning Robust Rewards with Adversarial Inverse Reinforcement Learningcited by 485
  9. 2023TD-MPC2: Scalable, Robust World Models for Continuous Controlcited by 482
  10. 2016Guided Cost Learning: Deep Inverse Optimal Control via Policy Optimizationcited by 400
  11. 2021Learning to Walk in Minutes Using Massively Parallel Deep Reinforcement Learningcited by 395
  12. 2022Temporal Difference Learning for Model Predictive Controlcited by 375
30 more
  1. 2023Orbit: A Unified Simulation Framework for Interactive Robot Learning Environmentscited by 239
  2. 2018Plan Online, Learn Offline: Efficient Learning and Exploration via Model-Based Controlcited by 104
  3. 2020Model-Based Inverse Reinforcement Learning from Visual Demonstrationscited by 52
  4. 2017Socially Compliant Navigation through Raw Depth Inputs with Generative Adversarial Imitation Learningcited by 50
  5. 2022Hierarchical Model-Based Imitation Learning for Planning in Autonomous Drivingcited by 43
  6. 2017Data-driven Planning via Imitation Learningcited by 36
  7. 2019Driving with Style: Inverse Reinforcement Learning in General-Purpose Planning for Automated Drivingcited by 30
  8. 2016Model-based Adversarial Imitation Learningcited by 29
  9. 2019An Online Learning Approach to Model Predictive Controlcited by 29
  10. 2023V-STRONG: Visual Self-Supervised Traversability Learning for Off-road Navigationcited by 28
  11. 2020Blending MPC & Value Function Approximation for Efficient Reinforcement Learningcited by 27
  12. 2021Model Predictive Actor-Critic: Accelerating Robot Skill Acquisition with Deep Reinforcement Learningcited by 25
  13. 2024Full-Order Sampling-Based MPC for Torque-Level Locomotion Control via Diffusion-Style Annealingcited by 20
  14. 2019MuSHR: A Low-Cost, Open-Source Robotic Racecar for Education and Researchcited by 17
  15. 2019Information Theoretic Model Predictive Q-Learningcited by 16
  16. 2023Learning Risk-Aware Costmaps via Inverse Reinforcement Learning for Off-Road Navigationcited by 15
  17. 2023Massively Scalable Inverse Reinforcement Learning in Google Mapscited by 11
  18. 2022Spatiotemporal Costmap Inference for MPC via Deep Inverse Reinforcement Learningcited by 10
  19. 2022Learning Sampling Distributions for Model Predictive Controlcited by 10
  20. 2021The Value of Planning for Infinite-Horizon Model Predictive Controlcited by 8
  21. 2023Model Predictive Control for Aggressive Driving Over Uneven Terraincited by 8
  22. 2024Distributional Successor Features Enable Zero-Shot Policy Optimizationcited by 8
  23. 2023Deep Model Predictive Optimizationcited by 7
  24. 2020Approximate Inverse Reinforcement Learning from Vision-based Imitation Learningcited by 6
  25. 2024Low Frequency Sampling in Model Predictive Path Integral Controlcited by 3
  26. 2024Legible and Proactive Robot Planning for Prosocial Human-Robot Interactionscited by 3
  27. 2024Dynamics Models in the Aggressive Off-Road Driving Regimecited by 3
  28. 2025Wheeled Lab: Modern Sim2Real for Low-cost, Open-source Wheeled Roboticscited by 3
  29. 2024Dynamic Non-Prehensile Object Transport via Model-Predictive Reinforcement Learningcited by 2
  30. 2024MPPI-Generic: A CUDA Library for Stochastic Trajectory Optimizationcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026Online World Modeling Enables Real-World Inverse Reinforcement Learning from Observationcited by 0
  2. 2026Discriminative Barrier Functions for Safe Adversarial Imitation Learning from Observationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2022Robust Learning from Observation with Model Misspecification
  2. 2022Imitating, Fast and Slow: Robust learning from demonstrations via decision-time planning
  3. 2018Reinforced Imitation: Sample Efficient Deep Reinforcement Learning for Map-less Navigation by Leveraging Prior Demonstrations
  4. 2019Dyna-AIL : Adversarial Imitation Learning by Planning
  5. 2022Imitation Learning by State-Only Distribution Matching
  6. 2018Deep Imitative Models for Flexible Inference, Planning, and Control
  7. 2022Plan Your Target and Learn Your Skills: Transferable State-Only Imitation Learning via Decoupled Policy Optimization
  8. 2022Robust Imitation of a Few Demonstrations with a Backwards Model