Distributed Multi-Robot Cooperation for Information Gathering under Communication Constraints

Alberto Viseras-Ruiz, Zhe Xu, and Luis Merino. Distributed Multi-Robot Cooperation for Information Gathering under Communication Constraints. In Proceedings of the IEEE International Conference on Robotics and Automation, ICRA, pp. 1267–1272, 2018.

Download

[PDF] [html] 

Abstract

Exploration is a key task in many robotic applications. The objective is to intelligently select the robot motions required to efficiently obtain a good reconstruction of a certain physical property, such as occupancy maps, wind fields or magnetic fields. This task can clearly benefit from multi-robot cooperation. Many exploration strategies employ discretization of the state and action spaces, which makes the strategies computationally intractable for robotic systems with complex dynamics. Moreover, many current algorithms cannot deal with additional restrictions like communication constraints. These problems are even more relevant when systems with multiple robots are considered. This paper presents an approach for multi-robot exploration that tackles these issues. It employs Gaussian processes to model the underlying process to be explored and sampling-based planners to plan paths in continuous domains considering kinematic constraints. Distributed decision-making for multi-robot exploration is achieved through the max-sum algorithm, employing information-theoretic utility functions for efficient exploration, while, at the same time, considering additional constraints for the robot team. Simulation results in a multi-robot exploration application show the effectiveness of the proposed approach..

BibTeX Entry

@INPROCEEDINGS{icra18multi,
  author = {Alberto Viseras-Ruiz and Zhe Xu and Luis Merino},
  title = {{Distributed Multi-Robot Cooperation for Information Gathering under Communication Constraints}},
  booktitle = ICRA,
  year = {2018},
  pages = {1267--1272},
  doi = {10.1109/ICRA.2018.8460846},
  abstract={Exploration is a key task in many robotic applications. The objective is to intelligently select the robot motions required to efficiently obtain a good reconstruction of a certain physical property, such as occupancy maps, wind fields or magnetic fields. This task can clearly benefit from multi-robot cooperation. Many exploration strategies employ discretization of the state and action spaces, which makes the strategies computationally intractable for robotic systems with complex dynamics. Moreover, many current algorithms cannot deal with additional restrictions like communication constraints. These problems are even more relevant when systems with multiple robots are considered. This paper presents an approach for multi-robot exploration that tackles these issues. It employs Gaussian processes to model the underlying process to be explored and sampling-based planners to plan paths in continuous domains considering kinematic constraints. Distributed decision-making for multi-robot exploration is achieved through the max-sum algorithm, employing information-theoretic utility functions for efficient exploration, while, at the same time, considering additional constraints for the robot team. Simulation results in a multi-robot exploration application show the effectiveness of the proposed approach..},
}

Generated by bib2html.pl (written by Patrick Riley ) on Mon Jan 09, 2023 18:40:36