A neural network approach to image-based navigation and localization

Tools

Short, Daniel L. and Lei, Tingjun and Liu, Lantao and Luo, Chaomin and Yang, Erfu; (2025) A neural network approach to image-based navigation and localization. In: 2025 International Joint Conference on Neural Networks (IJCNN). 2025 International Joint Conference on Neural Networks (IJCNN) . IEEE, ITA. ISBN 979-8-3315-1042-8 (https://doi.org/10.1109/IJCNN64981.2025.11227291)

[thumbnail of Short-etal-2025-A-neural-network-approach-to-image-based-navigation-and]
Preview
 Text. Filename: Short-etal-2025-A-neural-network-approach-to-image-based-navigation-and.pdf
Accepted Author Manuscript
License: Creative Commons Attribution 4.0 logo
Download (1MB)| Preview

Abstract

With the increasing use of autonomous robots and vehicles in unfamiliar environments, image-based navigation and localization have gained significant attention. Some advanced algorithms address robust and efficient global path planning. However, reliable local reactive navigation is crucial for obstacle avoidance in close proximity. This paper introduces a biologically inspired neural networks (BNN) model with a dynamic moving window method (DMWM) for local navigation, complemented by a bio-inspired Bat algorithm (BA) for global path planning. The BA utilizes visual features extracted from images using convolutional neural networks (CNNs) to generate paths for autonomous robots. This paper outlines the requirements for image-based navigation, addresses the BA’s principles and its suitability for global path planning, and details the development of the BNN with DMWM for local navigation. Finally, simulations and comparative studies validate the performance and reliability of the proposed methods.

ORCID iDs

Short, Daniel L., Lei, Tingjun, Liu, Lantao, Luo, Chaomin and Yang, Erfu ORCID logoORCID: https://orcid.org/0000-0003-1813-5950;
CORE (COnnecting REpositories)