Multi-robot dispatching in a geographically constrained environment

Authors:
Li-Che Chen;Yu Fu;Tien-Ruey Hsiang;Sheng-Luen Chung
Affiliations:
Department of Electrical Engineering, National Taiwan University of Science and Technology;Department of Electrical Engineering, National Taiwan University of Science and Technology;Department of Computer Science and Information Engineering, National Taiwan University of Science and Technology;Department of Electrical Engineering, National Taiwan University of Science and Technology
Venue:
CCDC'09 Proceedings of the 21st annual international conference on Chinese control and decision conference
Year:
2009

Citing 6
Cited 0

Data clustering: a review

ACM Computing Surveys (CSUR)
An empirical comparison of four initialization methods for the K-Means algorithm

Pattern Recognition Letters
An Efficient k-Means Clustering Algorithm: Analysis and Implementation

IEEE Transactions on Pattern Analysis and Machine Intelligence
Pattern Classification (2nd Edition)

Pattern Classification (2nd Edition)
Introduction to Autonomous Mobile Robots

Introduction to Autonomous Mobile Robots
A Neural Network Approach to Dynamic Task Assignment of Multirobots

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

We develop a task allocation approach using k-means algorithm and Steiner tree properties to dispatch a platoon of mobile robots in a constrained environment. Our approach increases task-execution efficiency of multiple robots by reducing the response time to each task. It clusters the request points based on the k-means algorithm, then pre-positions the robots at standby locations obtained by considering Steiner trees in their designated clusters. Robots travel along shortest paths to serve requests and have collision avoidance ability. A switching strategy for robots is also presented. By switching designated clusters, robots further shorten the total response time and balance their workloads implicitly. Thus, the total task execution time is reduced. The performance of our approach is evaluated through simulations.