让无人机导航更聪明:用语义感知的课程调度提升学习效率
SA-GCS: Semantic-Aware Gaussian Curriculum Scheduling for UAV Vision-Language Navigation
- 根据语义难度动态调整训练样本,实现由易到难的学习路径
- 在CityNav上超越强基线,收敛更快且性能更稳定
- 适合做无人机视觉语言导航的RL训练,尤其适合多规模模型
无人飞行器(UAV)视觉语言导航旨在基于自然语言指令,在复杂环境中准确定位目标并规划飞行路径,广泛应用于智能巡检、灾害救援和城市监控。近期视觉语言模型(VLMs)为该任务提供了强大的语义理解能力,而强化学习(RL)作为后训练策略展现出巨大潜力。然而,现有RL方法常因训练数据利用效率低、收敛慢、未充分考虑样本难度差异而受限。为此,我们提出**语义感知高斯课程调度(SA-GCS)**,将课程学习(CL)系统性地融入强化学习。SA-GCS采用语义感知难度估计器(SA-DE)量化训练样本复杂度,并通过高斯课程调度器(GCS)动态调整采样分布,实现从简单到复杂的平滑过渡。实验表明,该方法显著提升训练效率,加速收敛并增强整体性能。在CityNav基准测试中,SA-GCS在所有指标上均优于强基线,收敛更快且更稳定,且在不同规模模型间具有良好泛化能力,体现其鲁棒性与可扩展性。代码已公开。
原文摘要 · Abstract (English)
Unmanned Aerial Vehicle (UAV) Vision-Language Navigation (VLN) aims to enable agents to accurately localize targets and plan flight paths in complex environments based on natural language instructions, with broad applications in intelligent inspection, disaster rescue, and urban monitoring. Recent progress in Vision-Language Models (VLMs) has provided strong semantic understanding for this task, while reinforcement learning (RL) has emerged as a promising post-training strategy to further improve generalization. However, existing RL methods often suffer from inefficient use of training data, slow convergence, and insufficient consideration of the difficulty variation among training samples, which limits further performance improvement. To address these challenges, we propose \textbf{Semantic-Aware Gaussian Curriculum Scheduling (SA-GCS)}, a novel training framework that systematically integrates Curriculum Learning (CL) into RL. SA-GCS employs a Semantic-Aware Difficulty Estimator (SA-DE) to quantify the complexity of training samples and a Gaussian Curriculum Scheduler (GCS) to dynamically adjust the sampling distribution, enabling a smooth progression from easy to challenging tasks. This design significantly improves training efficiency, accelerates convergence, and enhances overall model performance. Extensive experiments on the CityNav benchmark demonstrate that SA-GCS consistently outperforms strong baselines across all metrics, achieves faster and more stable convergence, and generalizes well across models of different scales, highlighting its robustness and scalability. The implementation of our approach is publicly available.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品;中文卡片由大模型生成,请以原文为准。