CorrectNav: Self-Correction Flywheel Empowers Vision-Language-Action Navigation Model

TL;DR

CorrectNav introduces a self-correction flywheel paradigm that iteratively improves vision-language-action navigation models by leveraging their own error trajectories, leading to state-of-the-art success rates and robust real-world performance.

Contribution

The paper presents a novel self-correction flywheel training paradigm that uses error trajectories as data to enhance navigation models, a significant departure from traditional training methods.

Findings

Achieved new state-of-the-art success rates of 65.1% and 69.3% on R2R-CE and RxR-CE benchmarks.

Demonstrated superior real-world performance in obstacle avoidance and instruction following.

Validated the effectiveness of self-correction flywheel through multiple iterative improvements.

Abstract

Existing vision-and-language navigation models often deviate from the correct trajectory when executing instructions. However, these models lack effective error correction capability, hindering their recovery from errors. To address this challenge, we propose Self-correction Flywheel, a novel post-training paradigm. Instead of considering the model's error trajectories on the training set as a drawback, our paradigm emphasizes their significance as a valuable data source. We have developed a method to identify deviations in these error trajectories and devised innovative techniques to automatically generate self-correction data for perception and action. These self-correction data serve as fuel to power the model's continued training. The brilliance of our paradigm is revealed when we re-evaluate the model on the training set, uncovering new error trajectories. At this time, the self-correction flywheel begins to spin. Through multiple flywheel iterations, we progressively enhance our monocular RGB-based VLA navigation model CorrectNav. Experiments on R2R-CE and RxR-CE benchmarks show CorrectNav achieves new state-of-the-art success rates of 65.1% and 69.3%, surpassing prior best VLA navigation models by 8.2% and 16.4%. Real robot tests in various indoor and outdoor environments demonstrate \method's superior capability of error correction, dynamic obstacle avoidance, and long instruction following.