Saved in:
Bibliographic Details
Main Authors: Bakulin, Sergey, Akhtyamov, Timur, Fatykhov, Denis, Devchich, German, Ferrer, Gonzalo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2507.20892
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909709265534976
author Bakulin, Sergey
Akhtyamov, Timur
Fatykhov, Denis
Devchich, German
Ferrer, Gonzalo
author_facet Bakulin, Sergey
Akhtyamov, Timur
Fatykhov, Denis
Devchich, German
Ferrer, Gonzalo
contents This work proposes a novel hybrid approach for vision-only navigation of mobile robots, which combines advances of both deep learning approaches and classical model-based planning algorithms. Today, purely data-driven end-to-end models are dominant solutions to this problem. Despite advantages such as flexibility and adaptability, the requirement of a large amount of training data and limited interpretability are the main bottlenecks for their practical applications. To address these limitations, we propose a hierarchical system that utilizes recent advances in model predictive control, traversability estimation, visual place recognition, and pose estimation, employing topological graphs as a representation of the target environment. Using such a combination, we provide a scalable system with a higher level of interpretability compared to end-to-end approaches. Extensive real-world experiments show the efficiency of the proposed method.
format Preprint
id arxiv_https___arxiv_org_abs_2507_20892
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PixelNav: Towards Model-based Vision-Only Navigation with Topological Graphs
Bakulin, Sergey
Akhtyamov, Timur
Fatykhov, Denis
Devchich, German
Ferrer, Gonzalo
Robotics
This work proposes a novel hybrid approach for vision-only navigation of mobile robots, which combines advances of both deep learning approaches and classical model-based planning algorithms. Today, purely data-driven end-to-end models are dominant solutions to this problem. Despite advantages such as flexibility and adaptability, the requirement of a large amount of training data and limited interpretability are the main bottlenecks for their practical applications. To address these limitations, we propose a hierarchical system that utilizes recent advances in model predictive control, traversability estimation, visual place recognition, and pose estimation, employing topological graphs as a representation of the target environment. Using such a combination, we provide a scalable system with a higher level of interpretability compared to end-to-end approaches. Extensive real-world experiments show the efficiency of the proposed method.
title PixelNav: Towards Model-based Vision-Only Navigation with Topological Graphs
topic Robotics
url https://arxiv.org/abs/2507.20892