Spatial Relation Graph and Graph Convolutional Network for Object Goal Navigation
Robotics
2022-08-30 v1 Artificial Intelligence
Abstract
This paper describes a framework for the object-goal navigation task, which requires a robot to find and move to the closest instance of a target object class from a random starting position. The framework uses a history of robot trajectories to learn a Spatial Relational Graph (SRG) and Graph Convolutional Network (GCN)-based embeddings for the likelihood of proximity of different semantically-labeled regions and the occurrence of different object classes in these regions. To locate a target object instance during evaluation, the robot uses Bayesian inference and the SRG to estimate the visible regions, and uses the learned GCN embeddings to rank visible regions and select the region to explore next.
Cite
@article{arxiv.2208.13031,
title = {Spatial Relation Graph and Graph Convolutional Network for Object Goal Navigation},
author = {D. A. Sasi Kiran and Kritika Anand and Chaitanya Kharyal and Gulshan Kumar and Nandiraju Gireesh and Snehasis Banerjee and Ruddra dev Roychoudhury and Mohan Sridharan and Brojeshwar Bhowmick and Madhava Krishna},
journal= {arXiv preprint arXiv:2208.13031},
year = {2022}
}
Comments
CASE 2022 paper