mlsanityData to 14 Oct 2026

Learning Robot Manipulation from Cross-Morphology Demonstration

arXiv 2304.03833 ↗First posted 7 Apr 2023cs.RO cs.LGcited by 8

On the map: Offline RL and imitation learning from data → Robot manipulation imitation learning

Abstract (arXiv)

Some Learning from Demonstrations (LfD) methods handle small mismatches in the action spaces of the teacher and student. Here we address the case where the teacher's morphology is substantially different from that of the student. Our framework, Morphological Adaptation in Imitation Learning (MAIL), bridges this gap allowing us to train an agent from demonstrations by other agents with significantly different morphologies. MAIL learns from suboptimal demonstrations, so long as they provide textitsome guidance towards a desired solution. We demonstrate MAIL on manipulation tasks with rigid and deformable objects including 3D cloth manipulation interacting with rigid obstacles. We train a visual control policy for a robot with one end-effector using demonstrations from a simulated agent with two end-effectors. MAIL shows up to 24% improvement in a normalized performance metric over LfD and non-LfD baselines. It is deployed to a real Franka Panda robot, handles multiple variations in properties for objects (size, rotation, translation), and cloth-specific properties (color, thickness, size, material). An overview is on https://uscresl.github.io/mail .

Key facts. Cited by 8 papers in the map (data through 14 Oct 2026). Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data. Builds on 37 of its 55 references that are in the map. First posted on arXiv 7 Apr 2023.

Builds on

37 of 55 references.

Closest ancestors · 5

Same direction, newest first

  1. 2022Learning Deformable Object Manipulation from Expert Demonstrations20 Jul
  2. 2021Distilling Motion Planner Augmented Policies into Visual Control Policies for Robot Manipulation11 Nov
  3. 2021XIRL: Cross-embodiment Inverse Reinforcement Learning7 Jun
  4. 2020State-Only Imitation Learning for Dexterous Manipulation7 Apr
  5. 2018Zero-Shot Visual Imitation23 Apr

Other references · 32

  1. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  2. 2022RT-1: Robotics Transformer for Real-World Control at Scalecited by 2,061
  3. 2020Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problemscited by 1,920
  4. 2020Conservative Q-Learning for Offline Reinforcement Learningcited by 1,809
  5. 2016Generative Adversarial Imitation Learningcited by 1,597
  6. 2018Learning Latent Dynamics for Planning from Pixelscited by 1,264
  7. 2020Learning to Simulate Complex Physics with Graph Networkscited by 880
  8. 2020CURL: Contrastive Unsupervised Representations for Reinforcement Learningcited by 760
  9. 2017Learning Complex Dexterous Manipulation with Deep Reinforcement Learning and Demonstrationscited by 715
  10. 2020AWAC: Accelerating Online Reinforcement Learning with Offline Datasetscited by 671
  11. 2021Perceiver IO: A General Architecture for Structured Inputs & Outputscited by 629
  12. 2020Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixelscited by 585
20 more
  1. 2018Behavioral Cloning from Observationcited by 492
  2. 2022DayDreamer: World Models for Physical Robot Learningcited by 400
  3. 2017Leveraging Demonstrations for Deep Reinforcement Learning on Robotics Problems with Sparse Rewardscited by 319
  4. 2021Bridging Offline Reinforcement Learning and Imitation Learning: A Tale of Pessimismcited by 279
  5. 2017One-Shot Imitation Learningcited by 233
  6. 2017One-Shot Visual Imitation Learning via Meta-Learningcited by 228
  7. 2017DART: Noise Injection for Robust Imitation Learningcited by 191
  8. 2018Generative Adversarial Imitation from Observationcited by 184
  9. 2018Sim-to-Real Reinforcement Learning for Deformable Object Manipulationcited by 131
  10. 2020SoftGym: Benchmarking Deep Reinforcement Learning for Deformable Object Manipulationcited by 121
  11. 2019AVID: Learning Multi-Stage Tasks via Pixel-Level Translation of Human Videoscited by 105
  12. 2019Recent Advances in Imitation Learning from Observationcited by 93
  13. 2021Demonstration-Guided Reinforcement Learning with Learned Skillscited by 76
  14. 2020Super-Human Performance in Gran Turismo Sport Using Deep Reinforcement Learningcited by 65
  15. 2021Learning Visible Connectivity Dynamics for Cloth Smoothingcited by 62
  16. 2022Mesh-based Dynamics with Occlusion Reasoning for Cloth Manipulationcited by 34
  17. 2023LS-IQ: Implicit Reward Regularization for Inverse Reinforcement Learningcited by 29
  18. 2021Trajectory Optimization for Manipulation of Deformable Objects: Assembly of Belt Drive Unitscited by 11
  19. 2021Learning Visual Feedback Control for Dynamic Cloth Foldingcited by 10
  20. 2021Learning Periodic Tasks from Human Demonstrationscited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 6; 2025: 1; 2026: 1.

8 papers.

  1. 2024Scaling Cross-Embodied Learning: One Policy for Manipulation, Navigation, Locomotion and Aviationcited by 114
  2. 2024Cross-Embodiment Dexterous Grasping with Reinforcement Learningcited by 27
  3. 2024VoxAct-B: Voxel-Based Acting and Stabilizing Policy for Bimanual Manipulationcited by 21
  4. 2024A Comprehensive Survey of Cross-Domain Policy Transfer for Embodied Agentscited by 16
  5. 2025D-CODA: Diffusion for Coordinated Dual-Arm Data Augmentationcited by 5
  6. 2026SkillVLA: Tackling Combinatorial Diversity in Dual-Arm Manipulation via Skill Reusecited by 5
  7. 2024Knowledge Transfer for Cross-Domain Reinforcement Learning: A Systematic Reviewcited by 2
  8. 2024Reinforcement Learning from Wild Animal Videoscited by 2

Similar, not cited

Close in wording, no citation either way

  1. 2021Coarse-to-Fine Imitation Learning: Robot Manipulation from a Single Demonstration
  2. 2022Augmentation for Learning From Demonstration with Environmental Constraints
  3. 2026Data Analogies Enable Efficient Cross-Embodiment Transfer
  4. 2023Teach a Robot to FISH: Versatile Imitation from One Minute of Demonstrations
  5. 2022Eliciting Compatible Demonstrations for Multi-Human Imitation Learning
  6. 2026Human-to-Robot Interaction: Learning from Video Demonstration for Robot Imitation
  7. 2025Training Cross-Morphology Embodied AI Agents: From Practical Challenges to Theoretical Foundations
  8. 2024A Dual Approach to Imitation Learning from Observations with Offline Datasets