数据集

模拟器

VLN

综述

Awesome Vision–Language–Action Models for Autonomous Driving

GitHub

VLN-VER

似乎没有使用大模型

https://github.com/DefaultRui/VLN-VER

cvpr 2024

https://arxiv.org/pdf/2403.14158

https://blog.csdn.net/qq_35831906/article/details/135881625

VER能够预测3D占用、房间布局和3D框。

SG-Nav

https://zhuanlan.zhihu.com/p/909651478

分层 3D 场景图

增量式更新与剪枝

分层链式思维提示

基于图的重感知机制

UniGoal

https://github.com/bagh2178/UniGoal

VLMaps

https://github.com/vlmaps/vlmaps

https://zhuanlan.zhihu.com/p/1899511479981548204

Mapnav

https://zhuanlan.zhihu.com/p/28340277990

https://arxiv.org/pdf/2502.13451

https://zhuanlan.zhihu.com/p/26160636617

设计一种新颖的内存表征方法来有效替代传统的历史帧变得非常重要

Zero-shot Object-Centric Instruction Following: Integrating Foundation Models with Traditional Navigation

https://sonia-raychaudhuri.github.io/nlslam/

GridMM: Grid Memory Map for Vision-and-Language Navigation

Navid

Semantic Map

---

Semantic Map 综述

ASHiTA

Automatic Scene-grounded HIerarchical Task Analysis

https://arxiv.org/pdf/2504.06553

S-Graphs+: Real-time Localization and Mapping leveraging Hierarchical Representations

https://github.com/snt-arg/s_graphs_docker

https://arxiv.org/pdf/2212.11770

CURB-SG

https://github.com/robot-learning-freiburg/CURB-SG/tree/main

https://arxiv.org/pdf/2309.06635

https://github.com/koide3/hdl_graph_slam

HOV-SG (RSS 2024)

https://github.com/hovsg/HOV-SG

https://blog.csdn.net/qq_41204464/article/details/149445104