mlsanityData to 7 Oct 2026

CAD2RL: Real Single-Image Flight without a Single Real Image

arXiv 1611.04201 ↗First posted 13 Nov 2016cs.LG cs.CV cs.ROcited by 227

Not on the map. Why?

Cited by 21 in Offline RL and imitation learning from data, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Deep reinforcement learning has emerged as a promising and powerful technique for automatically acquiring control policies that can process raw sensory inputs, such as images, and perform complex behaviors. However, extending deep RL to real-world robotic tasks has proven challenging, particularly in safety-critical domains such as autonomous flight, where a trial-and-error learning process is often impractical. In this paper, we explore the following question: can we train vision-based navigation policies entirely in simulation, and then transfer them into the real world to achieve real-world flight without a single real training image? We propose a learning method that we call CAD2RL, which can be used to perform collision-free indoor flight in the real world while being trained entirely on 3D CAD models. Our method uses single RGB images from a monocular camera, without needing to explicitly reconstruct the 3D geometry of the environment or perform explicit motion planning. Our learned collision avoidance policy is represented by a deep convolutional neural network that directly processes raw monocular images and outputs velocity commands. This policy is trained entirely on simulated images, with a Monte Carlo policy evaluation algorithm that directly optimizes the network's ability to produce collision-free flight. By highly randomizing the rendering settings for our simulated training set, we show that we can train a policy that generalizes to the real world, without requiring the simulator to be particularly realistic or high-fidelity. We evaluate our method by flying a real quadrotor through indoor environments, and further evaluate the design choices in our simulator through a series of ablation studies on depth prediction. For supplementary video see: https://youtu.be/nXBWmzFrj5s

Key facts. Cited by 227 papers in the map (data through 7 Oct 2026). First posted on arXiv 13 Nov 2016.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2018: 1; 2019: 3; 2020: 3; 2021: 54; 2022: 50; 2023: 34; 2024: 23; 2025: 33; 2026: 26.

227 papers.

Sort:

By area:

  1. 2020Offline Reinforcement Learning: Tutorial, Review, and Perspectives on Open Problemscited by 1,873
  2. 2021Safe Learning in Robotics: From Learning-Based Control to Safe Reinforcement Learningcited by 337
  3. 2021A Survey of Embodied AI: From Simulators to Research Taskscited by 256
  4. 2021How to Train Your Robot with Deep Reinforcement Learning; Lessons We've Learnedcited by 201
  5. 2021Maximum Entropy RL (Provably) Solves Some Robust RL Problemscited by 169
  6. 2024Agent AI: Surveying the Horizons of Multimodal Interactioncited by 167
  7. 2022Evolving Curricula with Regret-Based Environment Designcited by 166
  8. 2021A Survey of Zero-shot Generalisation in Deep Reinforcement Learningcited by 164
  9. 2021FSDR: Frequency Space Domain Randomization for Domain Generalizationcited by 163
  10. 2024Deep Reinforcement Learning for Robotics: A Survey of Real-World Successescited by 153
  11. 2023Real-World Humanoid Locomotion with Reinforcement Learningcited by 132
  12. 2021Why Generalization in RL is Difficult: Epistemic POMDPs and Implicit Partial Observabilitycited by 102
38 more
  1. 2021Learning High-Speed Flight in the Wildcited by 98
  2. 2021The Distracting Control Suite -- A Challenging Benchmark for Reinforcement Learning from Pixelscited by 97
  3. 2021Reinforcement Learning for Robust Parameterized Locomotion Control of Bipedal Robotscited by 95
  4. 2020State-Only Imitation Learning for Dexterous Manipulationcited by 90
  5. 2025NoisyRollout: Reinforcing Visual Reasoning with Data Augmentationcited by 86
  6. 2021Q-attention: Enabling Efficient Learning for Vision-based Robotic Manipulationcited by 85
  7. 2022A Walk in the Park: Learning to Walk in 20 Minutes With Model-Free Reinforcement Learningcited by 76
  8. 2019Third-Person Visual Imitation Learning via Decoupled Hierarchical Controllercited by 73
  9. 2021Asymmetric self-play for automatic goal discovery in robotic manipulationcited by 71
  10. 2021Vision-Language Navigation with Random Environmental Mixupcited by 70
  11. 2021Beyond Pick-and-Place: Tackling Robotic Stacking of Diverse Shapescited by 70
  12. 2022CACTI: A Framework for Scalable Multi-Task Multi-Scene Visual Imitation Learningcited by 70
  13. 2021Understanding Domain Randomization for Sim-to-real Transfercited by 69
  14. 2022VeLO: Training Versatile Learned Optimizers by Scaling Upcited by 57
  15. 2021Semantics for Robotic Mapping, Perception and Interaction: A Surveycited by 56
  16. 2021Robot Learning from Randomized Simulations: A Reviewcited by 56
  17. 2023How Simulation Helps Autonomous Driving:A Survey of Sim2real, Digital Twins, and Parallel Intelligencecited by 55
  18. 2022Holo-Dex: Teaching Dexterity with Immersive Mixed Realitycited by 52
  19. 2021Auto-Tuned Sim-to-Real Transfercited by 45
  20. 2019Deep Imitation Learning of Sequential Fabric Smoothing From an Algorithmic Supervisorcited by 44
  21. 2022NeRF2Real: Sim2real Transfer of Vision-guided Bipedal Motion Skills using Neural Radiance Fieldscited by 41
  22. 2021Catch Me If You Hear Me: Audio-Visual Navigation in Complex Unmapped Environments with Moving Soundscited by 40
  23. 2022Deep Surrogate Assisted Generation of Environmentscited by 40
  24. 2022Adapting Rapid Motor Adaptation for Bipedal Robotscited by 39
  25. 2023What Makes Pre-Trained Visual Representations Successful for Robust Manipulation?cited by 38
  26. 2023A Finite Sample Complexity Bound for Distributionally Robust Q-learningcited by 35
  27. 2021Hierarchical Object-to-Zone Graph for Object Navigationcited by 34
  28. 2022Environment Generation for Zero-Shot Compositional Reinforcement Learningcited by 32
  29. 2022Trustworthy Reinforcement Learning Against Intrinsic Vulnerabilities: Robustness, Safety, and Generalizabilitycited by 32
  30. 2021Learning a State Representation and Navigation in Cluttered and Dynamic Environmentscited by 30
  31. 2021Unsupervised Visual Attention and Invariance for Reinforcement Learningcited by 30
  32. 2023MAESTRO: Open-Ended Environment Design for Multi-Agent Reinforcement Learningcited by 30
  33. 2025Vid2Sim: Realistic and Interactive Simulation from Video for Urban Navigationcited by 30
  34. 2022RISP: Rendering-Invariant State Predictor with Differentiable Simulation and Rendering for Cross-Domain Parameter Estimationcited by 29
  35. 2018Following High-level Navigation Instructions on a Simulated Quadcopter with Imitation Learningcited by 27
  36. 2022Imitate and Repurpose: Learning Reusable Robot Movement Skills From Human and Animal Behaviorscited by 27
  37. 2024ASID: Active Exploration for System Identification in Robotic Manipulationcited by 27
  38. 2023GINA-3D: Learning to Generate Implicit Neural Assets in the Wildcited by 25

50 most cited of 227 All 227 (JSON)