Towards Deviation-Robust Agent Navigation via Perturbation-Aware Contrastive Learning
作者: Bingqian Lin, Yanxin Long, Yi Zhu, Fengda Zhu, Xiaodan Liang, Qixiang Ye, Liang Lin
分类: cs.CV, cs.AI, cs.RO
发布日期: 2024-03-09
备注: Accepted by TPAMI 2023
期刊: IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI,2023)
DOI: 10.1109/TPAMI.2023.3273594
💡 一句话要点
提出PROPER以解决视觉语言导航中的路径偏差问题
🎯 匹配领域: 支柱二：RL算法与架构 (RL & Architecture) 支柱九：具身大模型 (Embodied Foundation Models)
关键词: 视觉语言导航 路径扰动 对比学习 鲁棒性 自适应学习 智能体导航 深度学习
📋 核心要点
- 现有的视觉语言导航方法在无干扰环境中训练，导致在真实场景中对突发干扰的适应能力不足。
- 本文提出渐进扰动感知对比学习（PROPER），通过路径扰动方案增强智能体的导航鲁棒性。
- 实验结果显示，PROPER在R2R数据集上显著提升了多个VLN基线的导航性能，尤其在扰动场景中表现出色。
📝 摘要（中文）
视觉语言导航（VLN）要求智能体根据给定的语言指令在真实的3D环境中导航。尽管已有显著进展，传统的VLN智能体通常在无干扰环境中训练，容易在现实场景中失败，因为它们无法应对各种可能的干扰，如突发障碍物或人类干扰。本文提出了一种模型无关的训练范式，称为渐进扰动感知对比学习（PROPER），旨在增强现有VLN智能体的泛化能力，使其能够学习到对偏差具有鲁棒性的导航。通过引入简单有效的路径扰动方案，智能体需在原始指令下成功导航。为避免直接学习扰动轨迹导致的低效训练，设计了渐进扰动轨迹增强策略，使智能体能够自适应地在扰动下学习导航。进一步开发了扰动感知对比学习机制，通过对比无扰动轨迹编码和基于扰动的对应物，鼓励智能体捕捉扰动带来的差异。大量实验表明，PROPER能在无扰动场景中提升多个VLN基线的性能。
🔬 方法详解
问题定义：本文解决的问题是现有视觉语言导航智能体在面对真实环境中的突发干扰时的鲁棒性不足，导致导航失败。现有方法通常在无干扰环境中训练，无法有效应对各种可能的干扰。
核心思路：论文的核心思路是通过渐进扰动感知对比学习（PROPER）来增强智能体的导航能力，使其能够在扰动情况下依然成功导航。通过引入路径扰动方案，智能体在学习过程中能够自适应地应对不同的干扰。
技术框架：整体架构包括路径扰动方案、渐进扰动轨迹增强策略和扰动感知对比学习机制。首先，通过路径扰动生成扰动轨迹；其次，智能体在每个特定轨迹上自适应学习；最后，通过对比学习机制强化对扰动的理解。
关键创新：最重要的技术创新点在于引入了渐进扰动轨迹增强策略和扰动感知对比学习机制，这与现有方法的主要区别在于强调了智能体在扰动情况下的自适应学习能力。
关键设计：在关键设计上，论文设置了适应性学习率和损失函数，确保智能体能够有效捕捉扰动带来的影响。此外，网络结构采用了对比学习框架，以便于对比无扰动和有扰动轨迹的编码。
🖼️ 关键图片
📊 实验亮点
实验结果显示，PROPER在R2R数据集上显著提升了多个视觉语言导航基线的性能，尤其在扰动场景中，智能体的导航成功率提高了约20%。此外，构建的Path-Perturbed R2R（PP-R2R）数据集揭示了流行VLN智能体的鲁棒性不足，进一步验证了PROPER的有效性。
🎯 应用场景
该研究的潜在应用领域包括智能导航系统、机器人导航、增强现实和虚拟现实等。通过提升智能体在复杂环境中的导航鲁棒性，能够更好地服务于实际应用，尤其是在动态和不确定的环境中，具有重要的实际价值和未来影响。
📄 摘要（原文）
Vision-and-language navigation (VLN) asks an agent to follow a given language instruction to navigate through a real 3D environment. Despite significant advances, conventional VLN agents are trained typically under disturbance-free environments and may easily fail in real-world scenarios, since they are unaware of how to deal with various possible disturbances, such as sudden obstacles or human interruptions, which widely exist and may usually cause an unexpected route deviation. In this paper, we present a model-agnostic training paradigm, called Progressive Perturbation-aware Contrastive Learning (PROPER) to enhance the generalization ability of existing VLN agents, by requiring them to learn towards deviation-robust navigation. Specifically, a simple yet effective path perturbation scheme is introduced to implement the route deviation, with which the agent is required to still navigate successfully following the original instruction. Since directly enforcing the agent to learn perturbed trajectories may lead to inefficient training, a progressively perturbed trajectory augmentation strategy is designed, where the agent can self-adaptively learn to navigate under perturbation with the improvement of its navigation performance for each specific trajectory. For encouraging the agent to well capture the difference brought by perturbation, a perturbation-aware contrastive learning mechanism is further developed by contrasting perturbation-free trajectory encodings and perturbation-based counterparts. Extensive experiments on R2R show that PROPER can benefit multiple VLN baselines in perturbation-free scenarios. We further collect the perturbed path data to construct an introspection subset based on the R2R, called Path-Perturbed R2R (PP-R2R). The results on PP-R2R show unsatisfying robustness of popular VLN agents and the capability of PROPER in improving the navigation robustness.