FLUX: Accelerating Cross-Embodiment Generative Navigation Policies via Rectified Flow and Static-to-Dynamic Learning

TL;DR

FLUX is a unified, flow-based navigation policy that efficiently handles static and dynamic tasks, improves inference speed, and transfers zero-shot across various robotic platforms.

Contribution

The paper introduces FLUX, a novel flow-based navigation method that unifies static and dynamic tasks and enhances efficiency and transferability.

Findings

FLUX improves inference speed by 47% over prior flow-based methods.

FLUX achieves state-of-the-art performance across six navigation tasks.

FLUX demonstrates zero-shot transfer on multiple robotic platforms.

Abstract

Autonomous navigation requires a broad spectrum of skills, from static goal-reaching to dynamic social traversal, yet evaluation remains fragmented across disparate protocols. We introduce DynBench, a dynamic navigation benchmark featuring physically valid crowd simulation. Combined with existing static protocols, it supports comprehensive evaluation across six fundamental navigation tasks. Within this framework, we propose FLUX, the first flow-based unified navigation policy. By linearizing probability flow, FLUX replaces iterative denoising with straight-line trajectories, improving per-step inference efficiency by 47% over prior flow-based methods and 29% over diffusion-based ones. Following a static-to-dynamic curriculum, FLUX initially establishes geometric priors and is subsequently refined through reinforcement learning in dynamic social environments. This regime not only strengthens socially-aware navigation but also enhances static task robustness by capturing recovery behaviors through stochastic action distributions. FLUX achieves state-of-the-art performance across all tasks and demonstrates zero-shot sim-to-real transfer on wheeled, quadrupedal, and humanoid platforms without any fine-tuning.