Real World Robotic Exploration using Deep Neural Networks Trained in Photorealistic Reconstructed Environments

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Ward, Isaac Ronald
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915497739550720
author Ward, Isaac Ronald
author_facet Ward, Isaac Ronald
contents In this work, an existing deep neural network approach for determining a robot's pose from visual information (RGB images) is modified, improving its localization performance without impacting its ease of training. Explicitly, the network's loss function is extended in a manner which intuitively combines the positional and rotational error in order to increase robustness to perceptual aliasing. An improvement in the localization accuracy for indoor scenes is observed: with decreases of up to 9.64% and 2.99% in the median positional and rotational error respectively, when compared to the unmodified network. Additionally, photogrammetry data is used to produce a pose-labelled dataset which allows the above model to be trained on a local environment, resulting in localization accuracies of 0.11m & 0.89 degrees. This trained model forms the basis of a navigation algorithm, which is tested in real-time on a TurtleBot (a wheeled robotic device). As such, this work introduces a full pipeline for creating a robust navigational algorithm for any given real world indoor scene; the only requirement being a collection of images from the scene, which can be captured in as little as 330 seconds of
format Preprint
id arxiv_https___arxiv_org_abs_2509_13342
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Real World Robotic Exploration using Deep Neural Networks Trained in Photorealistic Reconstructed Environments
Ward, Isaac Ronald
Robotics
Artificial Intelligence
In this work, an existing deep neural network approach for determining a robot's pose from visual information (RGB images) is modified, improving its localization performance without impacting its ease of training. Explicitly, the network's loss function is extended in a manner which intuitively combines the positional and rotational error in order to increase robustness to perceptual aliasing. An improvement in the localization accuracy for indoor scenes is observed: with decreases of up to 9.64% and 2.99% in the median positional and rotational error respectively, when compared to the unmodified network. Additionally, photogrammetry data is used to produce a pose-labelled dataset which allows the above model to be trained on a local environment, resulting in localization accuracies of 0.11m & 0.89 degrees. This trained model forms the basis of a navigation algorithm, which is tested in real-time on a TurtleBot (a wheeled robotic device). As such, this work introduces a full pipeline for creating a robust navigational algorithm for any given real world indoor scene; the only requirement being a collection of images from the scene, which can be captured in as little as 330 seconds of
title Real World Robotic Exploration using Deep Neural Networks Trained in Photorealistic Reconstructed Environments
topic Robotics
Artificial Intelligence
url https://arxiv.org/abs/2509.13342