Home /Research /Effective shepherding behaviours using multi-agent systems
SWARM

Effective shepherding behaviours using multi-agent systems

Kaoru Fujioka, Sakiko Hayashi

Year
2016
Citations
21

Abstract

The shepherding problem is to consider how a shepherd moves to herd sheep from their initial position to some target position. This problem arises in various situations in nature and has a number of useful applications such as in robotics and crowd control. Many simulations have been conducted to investigate herd and group behaviours. In contrast to these swarm analyses, fewer studies have investigated strategies for controlling a flock. Therefore, inspired by real sheepdog motion, the present study aims to determine how to effectively guide a herd with one shepherd taking into account cohesive power for sheep to gather. Our method is compared with previously proposed methods using computer simulations.

Keywords

Swarm roboticsPosition (finance)Artificial intelligenceHerdComputer scienceRoboticsContrast (vision)Swarm behaviourMachine learningRobot

Related papers

Browse all SWARM papers