Please use this identifier to cite or link to this item: https://hdl.handle.net/11000/31574

Computing the Traversability of the Environment by Means of Sparse Convolutional 3D Neural Networks


Thumbnail

View/Open:
 3-2023_ICINCO_Traversability_A_Santo (1).pdf

3,25 MB
Adobe PDF
Share:
Title:
Computing the Traversability of the Environment by Means of Sparse Convolutional 3D Neural Networks
Authors:
Santo López, Antonio  
Gil, Arturo  
Valiente, David
Ballesta, Mónica
Peidro, Adrian  
Editor:
INSTICC - Institute for Systems and Technologies of Information, Control and Communication
Department:
Departamentos de la UMH::Ingeniería de Sistemas y Automática
Issue Date:
2023-11
URI:
https://hdl.handle.net/11000/31574
Abstract:
The correct assessment of the environment in terms of traversability is strictly necessary during the navigation task in autonomous mobile robots. In particular, navigating along unknown, natural and unstructured environments requires techniques to select which areas can be traversed by the robot. In order to increase the autonomy of the system’s decisions, this paper proposes a method for the evaluation of 3D point clouds obtained by a LiDAR sensor in order to obtain the transitable areas, both in road and natural environments. Specifically, a trained sparse encoder-decoder configuration with rotation invariant features is proposed to replicate the input data by associating to each point the learned traversability features. Experimental results show the robustness and effectiveness of the proposed method in outdoor environments, improving the results of other approaches.
Keywords/Subjects:
Autonomous Mobile Robots
Artificial Intelligence
Neural Networks
Sparse Convolution
Point Clouds
Knowledge area:
CDU: Ciencias aplicadas: Ingeniería. Tecnología
Type of document:
application/pdf
Access rights:
info:eu-repo/semantics/openAccess
DOI:
https://doi.org/10.5220/0000168300003543
Appears in Collections:
Artículos Ingeniería de Sistemas y Automática



Creative Commons ???jsp.display-item.text9???