Shepherding Algorithm of Multi-mobile Robot System

This paper constructs a control system in which one highly mobile agent navigates autonomous multiple agents escaping from the agent according to nonlinear interaction. This study is motivated by the sheepdog system: a flock of thousands of sheep are controlled by only a few sheepdogs. Inspired by the sheepdog system, we have proposed a control law of a sheepdog to solve the problem of one sheep converging on a circular trajectory around a goal by one sheepdog, which is called the mobile control. In this paper, we extend this controller to applicable to a flock of sheep by treating multiple sheep as "a disc with a certain flock radius". This paper deals with the problem of navigating a flock of sheep to converge their centers of gravity on a circular trajectory around a goal position. For this problem, we propose a flock model of sheep and an extended mobile control designed by considering the suitable distance between a sheepdog and a center of the flock. Finally, we verified the validity of the proposed method through numerical simulation and robot demonstrations.

show abstract

Section: はじめにunclassified

Design of Mobile Control for Multiple Agents Inspired by Sheepdog Shepherding and its Verification

Tsunoda

Sueoka

Wada

et al. 2021

Transactions of ISCIE

View full text Add to dashboard Cite

show abstract

“…In order to select the optimal position for each shepherding agent in a multi shepherd system, the relative locations of the other shepherds as well as the sheep must be considered. Although it has been demonstrated that multi shepherd cooperation could naturally emerge [27], [28], a more common approach is to select a formation and assign each shepherding agent a position within the formation [20], [23], [24], [26], [29].…”

Section: Introductionmentioning

confidence: 99%

“…Another formation that has been used is caging, where the shepherds surround the flock, then coordinate their movements towards the goal. Bat-Erdene and Mandakh [26] utilized a square shaped cage formation. Interestingly, in their proposed solution the shepherding agents are heterogeneous with two distinct roles, "corner robot" and "sideline robot".…”

Section: Introductionmentioning

confidence: 99%

The Limits of Reactive Shepherding Approaches for Swarm Guidance

et al. 2020

View full text Add to dashboard Cite

“…The robotic shepherding problem, inspired by sheepdogs and sheep, can be defined as one or more shepherd robots try to guide a swarm of agents from an initial location to a goal location [1]. Shepherding is applicable to various fields in the real world, such as using robots to herd animals [2], [3], civil crowd control [4], preventing bird strikes near airports [5]- [7], facilitating communication between the unmanned ground vehicles and unmanned aerial vehicles [8].…”

Section: Introductionmentioning

confidence: 99%

Learning to Herd Agents Amongst Obstacles: Training Robust Shepherding Behaviors using Deep Reinforcement Learning

Zhi

Lien

2020

Preprint

View full text Add to dashboard Cite

Robotic shepherding problem considers the control and navigation of a group of coherent agents (e.g., a flock of bird or a fleet of drones) through the motion of an external robot, called shepherd. Machine learning based methods have successfully solved this problem in an empty environment with no obstacles. Rule-based methods, on the other hand, can handle more complex scenarios in which environments are cluttered with obstacles and allow multiple shepherds to work collaboratively. However, these rule-based methods are fragile due to the difficulty in defining a comprehensive set of rules that can handle all possible cases. To overcome these limitations, we propose the first known learning-based method that can herd agents amongst obstacles. By using deep reinforcement learning techniques combined with the probabilistic roadmaps, we train a shepherding model using noisy but controlled environmental and behavioral parameters. Our experimental results show that the proposed method is robust, namely, it is insensitive to the uncertainties originated from both environmental and behavioral models. Consequently, the proposed method has a higher success rate, shorter completion time and path length than the rule-based behavioral methods have. These advantages are particularly prominent in more challenging scenarios involving more difficult groups and strenuous passages.

show abstract

Shepherding Algorithm of Multi-mobile Robot System

Cited by 13 publications

References 4 publications

Design of Mobile Control for Multiple Agents Inspired by Sheepdog Shepherding and its Verification

Design of Mobile Control for Multiple Agents Inspired by Sheepdog Shepherding and its Verification

The Limits of Reactive Shepherding Approaches for Swarm Guidance

Learning to Herd Agents Amongst Obstacles: Training Robust Shepherding Behaviors using Deep Reinforcement Learning

Contact Info

Product

Resources

About