We propose a simple but effective modular approach MOPA (Modular ObjectNav with PointGoal agents) to systematically investigate the inherent modularity of the object navigation task in Embodied AI. MOPA consists of four modules: (a) an object detection module trained to identify objects from RGB images, (b) a map building module to build a semantic map of the observed objects, (c) an exploration module enabling the agent to explore the environment, and (d) a navigation module to move to identified target objects. We show that we can effectively reuse a pretrained PointGoal agent as the navigation model instead of learning to navigate from scratch, thus saving time and compute. We also compare various exploration strategies for MOPA and find that a simple uniform strategy significantly outperforms more advanced exploration methods.

翻译：我们提出一种简洁而高效的模块化方法MOPA（基于点目标智能体的模块化物体导航），系统性地探究具身智能中物体导航任务的内在模块化特性。MOPA包含四个模块：（a）物体检测模块，用于从RGB图像中识别物体；（b）地图构建模块，用于建立观测物体的语义地图；（c）探索模块，使智能体能够探索环境；（d）导航模块，用于移动至已识别的目标物体。研究表明，我们可有效复用预训练的点目标智能体作为导航模型，无需从头学习导航能力，从而节省时间与计算资源。此外，我们对比了MOPA的多种探索策略，发现简单的均匀策略显著优于更先进的探索方法。