MOPA: Modular Object Navigation with PointGoal Agents

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Raychaudhuri, Sonia, Campari, Tommaso, Jain, Unnat, Savva, Manolis, Chang, Angel X.
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911765711814656
author Raychaudhuri, Sonia
Campari, Tommaso
Jain, Unnat
Savva, Manolis
Chang, Angel X.
author_facet Raychaudhuri, Sonia
Campari, Tommaso
Jain, Unnat
Savva, Manolis
Chang, Angel X.
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2304_03696
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle MOPA: Modular Object Navigation with PointGoal Agents
Raychaudhuri, Sonia
Campari, Tommaso
Jain, Unnat
Savva, Manolis
Chang, Angel X.
Robotics
Computer Vision and Pattern Recognition
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.
title MOPA: Modular Object Navigation with PointGoal Agents
topic Robotics
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2304.03696