Empirical Analysis of Sim-and-Real Cotraining Of Diffusion Policies For Planar Pushing from Pixels

• URL: http://arxiv.org/abs/2503.22634v1
• Date: Fri, 28 Mar 2025 17:25:57 GMT
• Title: Empirical Analysis of Sim-and-Real Cotraining Of Diffusion Policies For Planar Pushing from Pixels
• Authors: Adam Wei, Abhinav Agarwal, Boyuan Chen, Rohan Bosworth, Nicholas Pfaff, Russ Tedrake,
• Abstract summary: Cotraining with demonstration data generated both in simulation and on real hardware has emerged as a powerful recipe to overcome the sim2real gap.<n>Performance gains scale with simulated data, but eventually plateau; real-world data increases this performance ceiling.<n>Perhaps surprisingly, having some visual domain gap actually helps the cotrained policy -- binary probes reveal that high-performing policies learn to distinguish simulated domains from real.
• Score: 18.858676073824515
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In imitation learning for robotics, cotraining with demonstration data generated both in simulation and on real hardware has emerged as a powerful recipe to overcome the sim2real gap. This work seeks to elucidate basic principles of this sim-and-real cotraining to help inform simulation design, sim-and-real dataset creation, and policy training. Focusing narrowly on the canonical task of planar pushing from camera inputs enabled us to be thorough in our study. These experiments confirm that cotraining with simulated data \emph{can} dramatically improve performance in real, especially when real data is limited. Performance gains scale with simulated data, but eventually plateau; real-world data increases this performance ceiling. The results also suggest that reducing the domain gap in physics may be more important than visual fidelity for non-prehensile manipulation tasks. Perhaps surprisingly, having some visual domain gap actually helps the cotrained policy -- binary probes reveal that high-performing policies learn to distinguish simulated domains from real. We conclude by investigating this nuance and mechanisms that facilitate positive transfer between sim-and-real. In total, our experiments span over 40 real-world policies (evaluated on 800+ trials) and 200 simulated policies (evaluated on 40,000+ trials).

Related papers

• Real-is-Sim: Bridging the Sim-to-Real Gap with a Dynamic Digital Twin for Real-World Robot Policy Evaluation [8.36634439225698]
  We propose real-is-sim, a behavior cloning framework that incorporates a dynamic digital twin throughout the policy development pipeline. We validate real-is-sim on the PushT manipulation task, demonstrating strong correlation between success rates obtained in the simulator and real-world evaluations.
  arXiv  Detail & Related papers  (2025-04-04T17:05:56Z)
• Sim-and-Real Co-Training: A Simple Recipe for Vision-Based Robotic Manipulation [40.96453435496208]
  We present a recipe for utilizing simulation data to solve vision-based robotic manipulation tasks. Using two domains--a robot arm and a humanoid--we demonstrate that simulation data can enhance real-world task performance by an average of 38%.
  arXiv  Detail & Related papers  (2025-03-31T17:39:38Z)
• Rapidly Adapting Policies to the Real World via Simulation-Guided Fine-Tuning [13.771418136861831]
  Physics simulators can generate vast data sets with broad coverage over states, actions, and environments.<n>Fine-tuning these policies with small real-world data sets is an appealing pathway for scaling robot learning.<n>This paper introduces the Simulation-Guided Fine-tuning (SGFT) framework, which demonstrates how to extract structural priors from physics simulators.
  arXiv  Detail & Related papers  (2025-02-04T20:40:44Z)
• Overcoming the Sim-to-Real Gap: Leveraging Simulation to Learn to Explore for Real-World RL [25.991354823569033]
  We show that in many regimes, while direct sim2real transfer may fail, we can utilize the simulator to learn a set of emphexploratory policies. In particular, in the setting of low-rank MDPs, we show that coupling these exploratory policies with simple, practical approaches. This is the first evidence that simulation transfer yields a provable gain in reinforcement learning in settings where direct sim2real transfer fails.
  arXiv  Detail & Related papers  (2024-10-26T19:12:27Z)
• Are NeRFs ready for autonomous driving? Towards closing the real-to-simulation gap [6.393953433174051]
  We propose a novel perspective for addressing the real-to-simulated data gap. We conduct the first large-scale investigation into the real-to-simulated data gap in an autonomous driving setting. Our results show notable improvements in model robustness to simulated data, even improving real-world performance in some cases.
  arXiv  Detail & Related papers  (2024-03-24T11:09:41Z)
• Benchmarking the Sim-to-Real Gap in Cloth Manipulation [10.530012817995656]
  We present a benchmark dataset to evaluate the sim-to-real gap in cloth manipulation. We use the dataset to evaluate the reality gap, computational time, and stability of four popular deformable object simulators.
  arXiv  Detail & Related papers  (2023-10-14T09:36:01Z)
• Sim2real Transfer Learning for Point Cloud Segmentation: An Industrial Application Case on Autonomous Disassembly [55.41644538483948]
  We present an industrial application case that uses sim2real transfer learning for point cloud data. We provide insights on how to generate and process synthetic point cloud data. A novel patch-based attention network is proposed additionally to tackle this problem.
  arXiv  Detail & Related papers  (2023-01-12T14:00:37Z)
• Simulation-Based Parallel Training [55.41644538483948]
  We present our ongoing work to design a training framework that alleviates those bottlenecks. It generates data in parallel with the training process. We present a strategy to mitigate this bias with a memory buffer.
  arXiv  Detail & Related papers  (2022-11-08T09:31:25Z)
• DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality [64.51295032956118]
  We train a policy that can perform robust dexterous manipulation on an anthropomorphic robot hand. Our work reaffirms the possibilities of sim-to-real transfer for dexterous manipulation in diverse kinds of hardware and simulator setups.
  arXiv  Detail & Related papers  (2022-10-25T01:51:36Z)
• Practical Imitation Learning in the Real World via Task Consistency Loss [18.827979446629296]
  This paper introduces a self-supervised loss that encourages sim and real alignment both at the feature and action-prediction levels. We achieve 80% success across ten seen and unseen scenes using only 16.2 hours of teleoperated demonstrations in sim and real.
  arXiv  Detail & Related papers  (2022-02-03T21:43:06Z)
• Reactive Long Horizon Task Execution via Visual Skill and Precondition Models [59.76233967614774]
  We describe an approach for sim-to-real training that can accomplish unseen robotic tasks using models learned in simulation to ground components of a simple task planner. We show an increase in success rate from 91.6% to 98% in simulation and from 10% to 80% success rate in the real-world as compared with naive baselines.
  arXiv  Detail & Related papers  (2020-11-17T15:24:01Z)
• Point Cloud Based Reinforcement Learning for Sim-to-Real and Partial Observability in Visual Navigation [62.22058066456076]
  Reinforcement Learning (RL) represents powerful tools to solve complex robotic tasks. RL does not work directly in the real-world, which is known as the sim-to-real transfer problem. We propose a method that learns on an observation space constructed by point clouds and environment randomization.
  arXiv  Detail & Related papers  (2020-07-27T17:46:59Z)
• SimAug: Learning Robust Representations from Simulation for Trajectory Prediction [78.91518036949918]
  We propose a novel approach to learn robust representation through augmenting the simulation training data. We show that SimAug achieves promising results on three real-world benchmarks using zero real training data.
  arXiv  Detail & Related papers  (2020-04-04T21:22:01Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.