OptiSight：弥合具身导航中的语义推理与几何控制

中文摘要 AI 辅助

自主室内导航既需要语义理解，又需要精确的几何控制。我们提出OptiSight，这是一种混合框架，通过有限状态思维链架构将视觉语言模型（VLM）推理与确定性视觉伺服相结合。Grounded-SAM定位开放词汇目标，而相机投影几何将视觉观测转换为导航命令，无需密集建图。VLM仅在关键决策点被查询，以降低计算开销，同时由几何控制处理连续导航。在AI Habitat中的实验表明，该方法在8GB VRAM预算内，可在包含避障和语义歧义的各种室内场景中实现可靠的零样本导航。源代码可在该https URL获取。

英文摘要

Autonomous indoor navigation requires both semantic understanding and precise geometric control. We propose OptiSight, a hybrid framework that combines Vision-Language Model reasoning with deterministic visual servoing through a finite-state Chain-of-Thought architecture. Grounded-SAM localizes open-vocabulary targets, while camera projection geometry converts visual observations into navigation commands without requiring dense mapping. The VLM is queried only at key decision points, reducing computational overhead while geometric control handles continuous navigation. Experiments in AI Habitat demonstrate reliable zero-shot navigation across diverse indoor scenarios, including obstacle avoidance and semantic ambiguity, while operating within an 8~GB VRAM budget. The source code is available at https://github.com/avanalperen/OptiSight-Python-Multimodal-CoT-for-Visual-Reasoning.

发表机构

• Institut de Robòtica i Informàtica Industrial (CSIC-UPC)（机器人学与工业信息研究所（西班牙国家研究委员会-加泰罗尼亚理工大学））

机构由 AI 辅助整理，请以论文原文为准。