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

2026-03-13

AI Summary

This work addresses the lack of a unified evaluation framework and generalizable strategies in existing autonomous navigation methods for tasks ranging from static goal reaching to dynamic social navigation. The authors propose FLUX, the first rectified flow–based unified navigation policy, which leverages curriculum learning from static to dynamic scenarios and incorporates reinforcement learning to enhance social interaction capabilities. Innovatively employing rectified flow modeling, FLUX replaces conventional iterative denoising with direct straight-trajectory generation, substantially improving inference efficiency. Evaluated on DynBench—a newly introduced dynamic navigation benchmark—FLUX achieves state-of-the-art performance across six tasks, with inference speeds 47% and 29% faster than prior flow-based and diffusion-based models, respectively. Notably, it demonstrates the first zero-shot sim-to-real transfer across wheeled, quadruped, and humanoid robots.

Technology Category

• Humans and AI: Human-Aware Planning and Behavior Prediction
• Natural Language Processing: Safety and Robustness
• Search and Optimization: Learning to Search

Application Category

• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

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.

Problem

• autonomous navigation
• dynamic navigation
• evaluation benchmark
• cross-embodiment
• unified policy

Innovation

• rectified flow
• static-to-dynamic learning
• unified navigation policy
• zero-shot sim-to-real transfer
• flow-based generative model