发表机构
Software Engineering Institute, East China Normal University; School of Geospatial Information, Information Engineering University; University of Shanghai for Science and Technology; Shanghai Jiao Tong University（华东师范大学软件工程学院; 信息工程大学地理空间信息学院; 上海理工大学; 上海交通大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对视觉导航持续学习难题，提出HyperDCM，通过场景图建模和记忆回放增强导航。利用视觉语言模型提取三元组，经R-GCN编码投影到双曲空间，采用动态聚类等策略。实验证明其在保留导航能力和泛化性上优于基线。
AI 中文摘要
视觉导航中的持续学习因灾难性遗忘以及适应多样且不断变化的环境的困难而仍具挑战性。为解决这些问题，我们提出双曲动态集群记忆（HyperDCM），这是一种结构感知记忆机制，通过场景图建模和有原则的记忆回放来增强基于扩散策略的导航。HyperDCM使用大型视觉语言模型从RGB观测中提取语义场景三元组，通过关系图卷积网络（R-GCN）将其编码为场景图嵌入，并将嵌入投影到双曲空间以增强连续导航中的结构可分离性和保留性。动态聚类和结构敏感更新策略选择代表性样本进行记忆回放，从而保持知识多样性并减轻灾难性遗忘。在多场景室内和室外数据集上的实验表明，与适用于扩散策略导航的代表性持续学习基线相比，HyperDCM在保留过去导航能力方面表现出色且泛化能力有所提高。
英文摘要
Continual learning in visual navigation remains challenging due to catastrophic forgetting and the difficulties associated with adapting to diverse and evolving environments. To address these issues, we propose Hyperbolic Dynamic Cluster Memory (HyperDCM), a structure-aware memory mechanism that enhances diffusion policy-based navigation through scene graph modeling and principled memory replay. HyperDCM extracts semantic scene triples from RGB observations using large vision-language models, encodes them into scene graph embeddings via a Relational Graph Convolutional Network (R-GCN), and projects the embeddings into hyperbolic space to enhance structural separability and retention in continual navigation. A dynamic clustering and structure-sensitive update strategy selects representative samples for memory replay, thereby preserving knowledge diversity and mitigating catastrophic forgetting. Experiments on multi-scene indoor and outdoor datasets demonstrate that HyperDCM achieves superior retention of past navigation capabilities and improved generalization compared to representative continual learning baselines adapted to diffusion policy navigation.
