Embodied-Navigator：指点-思考-记忆-对齐的高效导航
Original title: Embodied-Navigator: Point, Think, Memorize, and Align for Efficient Navigation
IndustryAI 75
Source: arXiv cs.ROPublish time unverified
arXiv:2608.17512v2 Announce Type: replace Abstract: Although Large Vision-Language Models (VLMs) have significantly advanced embodied navigation, their direct deployment remains challenging, as existing methods often force VLMs into unnatural action spaces that misalign with their 2D pre-training priors, compounded by rigid reasoning schedules and inefficient memory management. To overcome these limitations, we propose TAMP-Nav, a unified framework for efficient embodied navigation.