mlsanityData to 14 Oct 2026

RoamFlow: Reinforcement-Aligned One-Step Action MeanFlow Policy for Image-Goal Navigation

arXiv 2606.29934 ↗First posted 29 Jun 2026cs.ROcited by 0

On the map: Deep reinforcement learning methods → RL for robot navigation and UAV control

Abstract (arXiv)

Image-goal navigation is a key challenge in embodied robotics, where an agent must reach a target specified solely by a goal image. While existing reinforcement learning approaches map perceptual observations directly to actions, they struggle to model long-horizon dependencies, often leading to suboptimal trajectories. To address this limitation, we propose RoamFlow, a generative navigation framework that leverages MeanFlow to predict the average velocity field for trajectory synthesis, enabling efficient few-step generation and reducing inference latency. We further adopt a two-stage training strategy that combines expert imitation for stable initialization with reinforcement learning for task-specific policy refinement. Extensive experiments in both Habitat simulation and real-world robotic platforms demonstrate that RoamFlow achieves efficient inference while maintaining strong navigation performance under real-time constraints.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in RL for robot navigation and UAV control, one of 15 directions in Deep reinforcement learning methods. Builds on 33 of its 35 references that are in the map. First posted on arXiv 29 Jun 2026.

Builds on

33 of 35 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025SIGN: Safety-Aware Image-Goal Navigation for Autonomous Drones via Reinforcement Learning17 Aug

Other references · 32

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  4. 2017Proximal Policy Optimization Algorithmscited by 15,215
  5. 2022Flow Matching for Generative Modelingcited by 5,702
  6. 2018Soft Actor-Critic: Off-Policy Maximum Entropy Deep Reinforcement Learning with a Stochastic Actorcited by 4,885
  7. 2017FiLM: Visual Reasoning with a General Conditioning Layercited by 2,333
  8. 2023Consistency Modelscited by 1,822
  9. 2024Depth Anything V2cited by 1,422
  10. 2017Matterport3D: Learning from RGB-D Data in Indoor Environmentscited by 1,376
  11. 2019Habitat: A Platform for Embodied AI Researchcited by 1,191
  12. 2025Mean Flows for One-step Generative Modelingcited by 561
20 more
  1. 2018Gibson Env: Real-World Perception for Embodied Agentscited by 481
  2. 2016Target-driven Visual Navigation in Indoor Scenes using Deep Reinforcement Learningcited by 366
  3. 2024Diffusion Policy Policy Optimizationcited by 280
  4. 2023Denoising Diffusion Bridge Modelscited by 178
  5. 2024Consistency Policy: Accelerated Visuomotor Policies via Consistency Distillationcited by 166
  6. 2022Socially Compliant Navigation Dataset (SCAND): A Large-Scale Dataset of Demonstrations for Social Navigationcited by 127
  7. 2025ReinFlow: Fine-tuning Flow Matching Policy with Online Reinforcement Learningcited by 97
  8. 2024One-Step Diffusion Policy: Fast Visuomotor Policies via Diffusion Distillationcited by 89
  9. 2024One-Step Diffusion Distillation through Score Implicit Matchingcited by 71
  10. 2021Memory-Augmented Reinforcement Learning for Image-Goal Navigationcited by 66
  11. 2019Deep Visual MPC-Policy Learning for Navigationcited by 56
  12. 2023Knowledge Diffusion for Distillationcited by 53
  13. 2025NavDP: Learning Sim-to-Real Navigation Diffusion Policy with Privileged Information Guidancecited by 49
  14. 2025NaviDiffusor: Cost-Guided Diffusion Model for Visual Navigationcited by 25
  15. 2025Prior Does Matter: Visual Navigation via Denoising Diffusion Bridge Modelscited by 25
  16. 2024Variational Distillation of Diffusion Policies into Mixture of Expertscited by 15
  17. 2025FDPP: Fine-tune Diffusion Policy with Human Preferencecited by 15
  18. 2024FlowNav: Combining Flow Matching and Depth Priors for Efficient Navigationcited by 12
  19. 2025Score Distillation of Flow Matching Modelscited by 8
  20. 2025Fine-tuning Diffusion Policies with Backpropagation Through Diffusion Timestepscited by 7

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Transformers for Image-Goal Navigation
  2. 2026OGPO: One-Step Generative Policy Optimization for Real-Time Robot Control
  3. 2025MP1: MeanFlow Tames Policy Learning in 1-step for Robotic Manipulation
  4. 2026Action Agent: Agentic Video Generation Meets Flow-Constrained Diffusion
  5. 2023NoMaD: Goal Masked Diffusion Policies for Navigation and Exploration
  6. 2025EmbodiedDiffusion: End-to-End Traversability-Guided Visual Diffusion for Heterogeneous Robot Navigation
  7. 2026120 Minutes and a Laptop: Minimalist Image-goal Navigation via Unsupervised Exploration and Offline RL
  8. 2025Exploration and Adaptation in Non-Stationary Tasks with Diffusion Policies