.png){kind=icon}
.png){kind=icon}
Return
Collaborative Multi-MSA Multi-Target Tracking and Surveillance: a Divide & Conquer Method Using Region Allocation Trees
.png){kind=icon}
PRE
AI
DOI：10.1007/s10846-017-0499-4.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This paper presents a concurrent region decomposition and allocation algorithm that solves the multi-MSA coordination problem within the context of multi-target tracking and surveillance missions. Our collaboration approach achieves favorable computational characteristics, compared to its alternatives, by taking advantage of a data structure we named region allocation tree and the recursive processing strategy it allows. The region allocation tree data structure identifies the candidate regions, organizes information pertaining to tracking uncertainties and mobile sensor agent assignments, and allows for region decomposition and allocation simultaneously in a single depth-first sweep. Our collaboration approach, here, is used in conjunction with a Bayesian tracking algorithm-as the decision making is carried out in the belief space. Our contributions are also located within the wider discourse on multi-robot coordination. The simulation results demonstrate the effectiveness of our multi-MSA coordination approach.
Keywords:
Mobile sensor agents
Multi-robot systems
Task allocation
Region allocation
Target tracking
Area surveillance
Decision making
Bayesian tracking
Region allocation tree
Multi-MSA collaboration
Multi-UAV collaboration
Active information gathering
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
J
IF:
2.8
Papers:
3.8K
Citations:
6.9K