mlsanityData to 14 Oct 2026

Zero-Shot Transfer of Haptics-Based Object Insertion Policies

arXiv 2301.12587 ↗First posted 29 Jan 2023cs.RO cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

Humans naturally exploit haptic feedback during contact-rich tasks like loading a dishwasher or stocking a bookshelf. Current robotic systems focus on avoiding unexpected contact, often relying on strategically placed environment sensors. Recently, contact-exploiting manipulation policies have been trained in simulation and deployed on real robots. However, they require some form of real-world adaptation to bridge the sim-to-real gap, which might not be feasible in all scenarios. In this paper we train a contact-exploiting manipulation policy in simulation for the contact-rich household task of loading plates into a slotted holder, which transfers without any fine-tuning to the real robot. We investigate various factors necessary for this zero-shot transfer, like time delay modeling, memory representation, and domain randomization. Our policy transfers with minimal sim-to-real gap and significantly outperforms heuristic and learnt baselines. It also generalizes to plates of different sizes and weights. Demonstration videos and code are available at https://sites.google.com/view/compliant-object-insertion.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 24 of its 49 references that are in the map. First posted on arXiv 29 Jan 2023.

Builds on

24 of 49 references.

References · 24

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2018Soft Actor-Critic Algorithms and Applicationscited by 1,240
  3. 2018Learning Dexterous In-Hand Manipulationcited by 687
  4. 2020robosuite: A Modular Simulation Framework and Benchmark for Robot Learningcited by 511
  5. 2015Deep Recurrent Q-Learning for Partially Observable MDPscited by 465
  6. 2018Residual Reinforcement Learning for Robot Controlcited by 274
  7. 2017Reverse Curriculum Generation for Reinforcement Learningcited by 198
  8. 2018Making Sense of Vision and Touch: Self-Supervised Learning of Multimodal Representations for Contact-Rich Taskscited by 154
  9. 2021ContactOpt: Optimizing Contact to Improve Graspscited by 139
  10. 2014Reducing the Barrier to Entry of Complex Robotic Software: a MoveIt! Case Studycited by 89
  11. 2019Making Sense of Vision and Touch: Learning Multimodal Representations for Contact-Rich Taskscited by 84
  12. 2019Deep Reinforcement Learning for Industrial Insertion Tasks with Visual Inputs and Natural Rewardscited by 64
12 more
  1. 2021Tactile-RL for Insertion: Generalization to Objects of Unknown Geometrycited by 48
  2. 2017Deep Reinforcement Learning for High Precision Assembly Taskscited by 39
  3. 2021kPAM 2.0: Feedback Control for Category-Level Robotic Manipulationcited by 38
  4. 2018A Practical Approach to Insertion with Variable Socket Position Using Deep Reinforcement Learningcited by 36
  5. 2020Meta-Reinforcement Learning for Robotic Industrial Insertion Taskscited by 34
  6. 2020Variable Compliance Control for Robotic Peg-in-Hole Assembly: A Deep Reinforcement Learning Approachcited by 29
  7. 2021Reverb: A Framework For Experience Replaycited by 27
  8. 2022A System for Imitation Learning of Contact-Rich Bimanual Manipulation Policiescited by 26
  9. 2018A Framework for Robot Manipulation: Skill Formalism, Meta Learning and Adaptive Controlcited by 17
  10. 2015Occlusion-Aware Object Localization, Segmentation and Pose Estimationcited by 4
  11. 2022Prim-LAfD: A Framework to Learn and Adapt Primitive-Based Skills from Demonstrations for Insertion Taskscited by 4
  12. 2020Imitation Learning for High Precision Peg-in-Hole Taskscited by 3

Built on by

Cited by year

Citing papers per year: 2024: 2.

2 papers.

  1. 2024Symmetry-aware Reinforcement Learning for Robotic Assembly under Partial Observability with a Soft Wristcited by 9
  2. 2024GenCHiP: Generating Robot Policy Code for High-Precision and Contact-Rich Manipulation Taskscited by 9