DreamFlow: Local Navigation Beyond Observation via Conditional Flow Matching in the Latent Space

TL;DR

DreamFlow introduces a deep reinforcement learning framework with conditional flow matching in the latent space, enabling robots to predict unobserved obstacles and improve local navigation in cluttered environments.

Contribution

It proposes a novel CFM-based prediction module that extends perceptual horizon, allowing better obstacle prediction and navigation beyond the robot's immediate sensing.

Findings

Outperforms existing methods in simulation accuracy and navigation success.

Successfully validated on a quadrupedal robot in real-world cluttered environments.

Enhances local navigation by predicting unobserved environmental features.

Abstract

Local navigation in cluttered environments often suffers from dense obstacles and frequent local minima. Conventional local planners rely on heuristics and are prone to failure, while deep reinforcement learning(DRL)based approaches provide adaptability but are constrained by limited onboard sensing. These limitations lead to navigation failures because the robot cannot perceive structures outside its field of view. In this paper, we propose DreamFlow, a DRL-based local navigation framework that extends the robot's perceptual horizon through conditional flow matching(CFM). The proposed CFM based prediction module learns probabilistic mapping between local height map latent representation and broader spatial representation conditioned on navigation context. This enables the navigation policy to predict unobserved environmental features and proactively avoid potential local minima. Experimental results demonstrate that DreamFlow outperforms existing methods in terms of latent prediction accuracy and navigation performance in simulation. The proposed method was further validated in cluttered real world environments with a quadrupedal robot. The project page is available at https://dreamflow-icra.github.io.