Multi-Domain Informative Coverage Path Planning for a Hybrid Aerial Underwater Vehicle in Dynamic Environments
Hybrid aerial underwater vehicles (HAUV) are a new frontier for vehicles. They can operate both underwater and aerially, providing enormous potential for a wide range of scientific explorations. Informative path planning is essential to vehicle autonomy. However, covering an entire mission region is...
Guardado en:
Autores principales: | , , |
---|---|
Formato: | article |
Lenguaje: | EN |
Publicado: |
MDPI AG
2021
|
Materias: | |
Acceso en línea: | https://doaj.org/article/bbe2520059954f05b92068d8bcfa7a0b |
Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
id |
oai:doaj.org-article:bbe2520059954f05b92068d8bcfa7a0b |
---|---|
record_format |
dspace |
spelling |
oai:doaj.org-article:bbe2520059954f05b92068d8bcfa7a0b2021-11-25T18:12:14ZMulti-Domain Informative Coverage Path Planning for a Hybrid Aerial Underwater Vehicle in Dynamic Environments10.3390/machines91102782075-1702https://doaj.org/article/bbe2520059954f05b92068d8bcfa7a0b2021-11-01T00:00:00Zhttps://www.mdpi.com/2075-1702/9/11/278https://doaj.org/toc/2075-1702Hybrid aerial underwater vehicles (HAUV) are a new frontier for vehicles. They can operate both underwater and aerially, providing enormous potential for a wide range of scientific explorations. Informative path planning is essential to vehicle autonomy. However, covering an entire mission region is a challenge to HAUVs because of the possibility of a multidomain environment. This paper presents an informative trajectory planning framework for planning paths and generating trajectories for HAUVs performing multidomain missions in dynamic environments. We introduce the novel heuristic generalized extensive neighborhood search GLNS–k-means algorithm that uses k-means to cluster information into several sets; then through the heuristic GLNS algorithm, it searches the best path for visiting these points, subject to various constraints regarding path budgets and the motion capabilities of the HAUV. With this approach, the HAUV is capable of sampling and focusing on regions of interest. Our method provides a significantly more optimal trajectory (enabling collection of more information) than ant colony optimization (ACO) solutions. Moreover, we introduce an efficient online replanning scheme to adapt the trajectory according to the dynamic obstacles during the mission. The proposed replanning scheme based on KD tree enables significantly shorter computational times than the scapegoat tree methods.Xueyao LiangChunhu LiuZheng ZengMDPI AGarticleunderwater vehiclehybrid aerial underwater vehicleinformative path planningMechanical engineering and machineryTJ1-1570ENMachines, Vol 9, Iss 278, p 278 (2021) |
institution |
DOAJ |
collection |
DOAJ |
language |
EN |
topic |
underwater vehicle hybrid aerial underwater vehicle informative path planning Mechanical engineering and machinery TJ1-1570 |
spellingShingle |
underwater vehicle hybrid aerial underwater vehicle informative path planning Mechanical engineering and machinery TJ1-1570 Xueyao Liang Chunhu Liu Zheng Zeng Multi-Domain Informative Coverage Path Planning for a Hybrid Aerial Underwater Vehicle in Dynamic Environments |
description |
Hybrid aerial underwater vehicles (HAUV) are a new frontier for vehicles. They can operate both underwater and aerially, providing enormous potential for a wide range of scientific explorations. Informative path planning is essential to vehicle autonomy. However, covering an entire mission region is a challenge to HAUVs because of the possibility of a multidomain environment. This paper presents an informative trajectory planning framework for planning paths and generating trajectories for HAUVs performing multidomain missions in dynamic environments. We introduce the novel heuristic generalized extensive neighborhood search GLNS–k-means algorithm that uses k-means to cluster information into several sets; then through the heuristic GLNS algorithm, it searches the best path for visiting these points, subject to various constraints regarding path budgets and the motion capabilities of the HAUV. With this approach, the HAUV is capable of sampling and focusing on regions of interest. Our method provides a significantly more optimal trajectory (enabling collection of more information) than ant colony optimization (ACO) solutions. Moreover, we introduce an efficient online replanning scheme to adapt the trajectory according to the dynamic obstacles during the mission. The proposed replanning scheme based on KD tree enables significantly shorter computational times than the scapegoat tree methods. |
format |
article |
author |
Xueyao Liang Chunhu Liu Zheng Zeng |
author_facet |
Xueyao Liang Chunhu Liu Zheng Zeng |
author_sort |
Xueyao Liang |
title |
Multi-Domain Informative Coverage Path Planning for a Hybrid Aerial Underwater Vehicle in Dynamic Environments |
title_short |
Multi-Domain Informative Coverage Path Planning for a Hybrid Aerial Underwater Vehicle in Dynamic Environments |
title_full |
Multi-Domain Informative Coverage Path Planning for a Hybrid Aerial Underwater Vehicle in Dynamic Environments |
title_fullStr |
Multi-Domain Informative Coverage Path Planning for a Hybrid Aerial Underwater Vehicle in Dynamic Environments |
title_full_unstemmed |
Multi-Domain Informative Coverage Path Planning for a Hybrid Aerial Underwater Vehicle in Dynamic Environments |
title_sort |
multi-domain informative coverage path planning for a hybrid aerial underwater vehicle in dynamic environments |
publisher |
MDPI AG |
publishDate |
2021 |
url |
https://doaj.org/article/bbe2520059954f05b92068d8bcfa7a0b |
work_keys_str_mv |
AT xueyaoliang multidomaininformativecoveragepathplanningforahybridaerialunderwatervehicleindynamicenvironments AT chunhuliu multidomaininformativecoveragepathplanningforahybridaerialunderwatervehicleindynamicenvironments AT zhengzeng multidomaininformativecoveragepathplanningforahybridaerialunderwatervehicleindynamicenvironments |
_version_ |
1718411520873857024 |