Paper Reading AI Learner

Embodied Navigation Foundation Model

2025-09-15 16:52:43
Jiazhao Zhang, Anqi Li, Yunpeng Qi, Minghan Li, Jiahang Liu, Shaoan Wang, Haoran Liu, Gengze Zhou, Yuze Wu, Xingxing Li, Yuxin Fan, Wenjun Li, Zhibo Chen, Fei Gao, Qi Wu, Zhizheng Zhang, He Wang

Abstract

Navigation is a fundamental capability in embodied AI, representing the intelligence required to perceive and interact within physical environments following language instructions. Despite significant progress in large Vision-Language Models (VLMs), which exhibit remarkable zero-shot performance on general vision-language tasks, their generalization ability in embodied navigation remains largely confined to narrow task settings and embodiment-specific architectures. In this work, we introduce a cross-embodiment and cross-task Navigation Foundation Model (NavFoM), trained on eight million navigation samples that encompass quadrupeds, drones, wheeled robots, and vehicles, and spanning diverse tasks such as vision-and-language navigation, object searching, target tracking, and autonomous driving. NavFoM employs a unified architecture that processes multimodal navigation inputs from varying camera configurations and navigation horizons. To accommodate diverse camera setups and temporal horizons, NavFoM incorporates identifier tokens that embed camera view information of embodiments and the temporal context of tasks. Furthermore, to meet the demands of real-world deployment, NavFoM controls all observation tokens using a dynamically adjusted sampling strategy under a limited token length budget. Extensive evaluations on public benchmarks demonstrate that our model achieves state-of-the-art or highly competitive performance across multiple navigation tasks and embodiments without requiring task-specific fine-tuning. Additional real-world experiments further confirm the strong generalization capability and practical applicability of our approach.

Abstract (translated)

导航是具身人工智能(Embodied AI)中的一个基本能力,它代表了感知和在物理环境中根据语言指令进行互动所需的情报。尽管大型视觉-语言模型(VLMs)已经在一般的视觉-语言任务上展示了显著的零样本性能,但它们在具身导航中的泛化能力仍然主要局限于狭窄的任务设置和特定于具身化的架构。在这项工作中,我们引入了一个跨具身形式和跨任务的导航基础模型(NavFoM),该模型基于涵盖四足动物、无人机、轮式机器人和车辆等八个百万导航样本进行训练,并覆盖了诸如视觉-语言导航、物体搜索、目标跟踪及自主驾驶等多样化任务。NavFoM采用了一种统一架构,能够处理来自不同相机配置和导航范围的多模态导航输入。 为了适应多样化的相机设置和时间范围,NavFoM采用了标识令牌来嵌入具身形式的视图信息以及任务的时间背景。此外,为满足实际部署的需求,NavFoM在有限的标记长度预算下使用动态调整的采样策略来控制所有观察标记。在公开基准测试上的广泛评估表明,在无需特定任务微调的情况下,我们的模型在多个导航任务和具身形式上实现了最先进的或具有高度竞争力的表现。此外,真实世界的实验进一步证实了我们方法的强大泛化能力和实际应用性。

URL

https://arxiv.org/abs/2509.12129

PDF

https://arxiv.org/pdf/2509.12129.pdf


Tags
3D Action Action_Localization Action_Recognition Activity Adversarial Agent Attention Autonomous Bert Boundary_Detection Caption Chat Classification CNN Compressive_Sensing Contour Contrastive_Learning Deep_Learning Denoising Detection Dialog Diffusion Drone Dynamic_Memory_Network Edge_Detection Embedding Embodied Emotion Enhancement Face Face_Detection Face_Recognition Facial_Landmark Few-Shot Gait_Recognition GAN Gaze_Estimation Gesture Gradient_Descent Handwriting Human_Parsing Image_Caption Image_Classification Image_Compression Image_Enhancement Image_Generation Image_Matting Image_Retrieval Inference Inpainting Intelligent_Chip Knowledge Knowledge_Graph Language_Model LLM Matching Medical Memory_Networks Multi_Modal Multi_Task NAS NMT Object_Detection Object_Tracking OCR Ontology Optical_Character Optical_Flow Optimization Person_Re-identification Point_Cloud Portrait_Generation Pose Pose_Estimation Prediction QA Quantitative Quantitative_Finance Quantization Re-identification Recognition Recommendation Reconstruction Regularization Reinforcement_Learning Relation Relation_Extraction Represenation Represenation_Learning Restoration Review RNN Robot Salient Scene_Classification Scene_Generation Scene_Parsing Scene_Text Segmentation Self-Supervised Semantic_Instance_Segmentation Semantic_Segmentation Semi_Global Semi_Supervised Sence_graph Sentiment Sentiment_Classification Sketch SLAM Sparse Speech Speech_Recognition Style_Transfer Summarization Super_Resolution Surveillance Survey Text_Classification Text_Generation Time_Series Tracking Transfer_Learning Transformer Unsupervised Video_Caption Video_Classification Video_Indexing Video_Prediction Video_Retrieval Visual_Relation VQA Weakly_Supervised Zero-Shot