A Comprehensive Review of Shepherding as a Bio-Inspired Swarm-Robotics Guidance Approach

Long, Nathan K.; Sammut, Karl; Sgarioto, Daniel; Garratt, Matthew A.; Abbass, Hussein A.

doi:10.1109/tetci.2020.2992778

Cited by 77 publications

(54 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Influencing the swarm through leaders who contribute operator-biased parameter values to their neighbors provides a consensus preserving path of control that additionally benefits from a direct correspondence between input (heading and velocity in the case of flocking) and desired effect on the swarm. Herding (Long et al 2020) offers a bio-inspired alternative to leaders based on sheep dog behavior and repulsion. In herding, however, the operator cannot conveniently influence the entire swarm and in particular, for multiple controlled herders, must account for complex interactions to accomplish a desired effect.…”

Section: Swarmsmentioning

confidence: 99%

Human vs. Deep Neural Network Performance at a Leader Identification Task

Deka¹,

Sycara

Walker

et al. 2021

Proceedings of the Human Factors and Ergonomics Society Annual

View full text Add to dashboard Cite

Control of robotic swarms through control over a leader(s) has become the dominant approach to supervisory control over these largely autonomous systems. Resilience in the face of attrition is one of the primary advantages attributed to swarms yet the presence of leader(s) makes them vulnerable to decapitation. Algorithms which allow a swarm to hide its leader are a promising solution. We present a novel approach in which neural networks, NNs, trained in a graph neural network, GNN, replace conventional controllers making them more amenable to training. Swarms and an adversary intent of finding the leader were trained and tested in 4 phases: 1-swarm to follow leader, 2-adversary to recognize leader, 3-swarm to hide leader from adversary, and 4-swarm and adversary compete to hide and recognize the leader. While the NN adversary was more successful in identifying leaders without deception, humans did better in conditions in which the swarm was trained to hide its leader from the NN adversary. The study illustrates difficulties likely to emerge in arms races between machine learners and the potential role humans may play in moderating them.

show abstract

Section: Swarmsmentioning

confidence: 99%

Human vs. Deep Neural Network Performance at a Leader Identification Task

Deka¹,

Sycara

Walker

et al. 2021

Proceedings of the Human Factors and Ergonomics Society Annual

View full text Add to dashboard Cite

show abstract

“…One approach for guidance and control that started to gain attentions in recent years is shepherding. Long et al [7] surveys the shepherding literature and offers a comprehensive coverage of the topic.…”

Section: Introductionmentioning

confidence: 99%

“…The shepherding problem definition is generic enough that it has been transferred to many applications including crowd control [9], rerouting and evacuation [10], keeping birds away from airports [11], collecting oil spills [12], and herding living cells to perform collective migrations [13]. A recent survey on the topic lists more domain of applications [7]. More recently, Strömbom et al [8] adopted a similar model to Miki and Nakamura, but with different naming conventions.…”

Section: Introductionmentioning

confidence: 99%

The Limits of Reactive Shepherding Approaches for Swarm Guidance

et al. 2020

Self Cite

View full text Add to dashboard Cite

“…The swarm shepherding problem has been researched in various contexts by many authors [1]. The problem has seen extensive use in different domains of applications including biological immune systems [2], horse harem groups [3], birds [4], and sheep [5], ground [6] and aerial robotics [7], [8].…”

Section: Introductionmentioning

confidence: 99%

Tracking Footprints in a Swarm: Information-Theoretic and Spatial Centre of Influence Measures

Hepworth¹,

yaxley²,

Baxter³

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

<div><div><div><p>Boids (Bird-oids) is a widely used model to mimic the behaviour of birds. Shoids (Sheep-oids) rely on the same boids rules with the addition of a repulsive force away from a sheepdog to model predation risk in predator-prey dynamic. Previous work assumed homogeneous shoids. Real-world observations on sheep show non-homogeneous responses to the presence of a herding agent. We present a portfolio of information-theoretic and spatial indicators to track the footprints of shoid with different parameters from the remainder of the shoid flock. The portfolio is named the Centre of Influence to indicate that the aim is to identify the influential shoids with the highest impact on flock dynamics. We use both synthetic simulation-driven data and measurements collected from actual sheep herding trial by an Unmanned Aerial Vehicle (UAV) to validate the proposed measures. The resultant footprints will allow us in our future research to design more efficient control strategies for the UAV to improve the herding of sheep, by polarising the attention of the machine learning algorithm on those Shoids with influence footprints to drive the flock.</p></div></div></div>

show abstract

A Comprehensive Review of Shepherding as a Bio-Inspired Swarm-Robotics Guidance Approach

Cited by 77 publications

References 50 publications

Human vs. Deep Neural Network Performance at a Leader Identification Task

Human vs. Deep Neural Network Performance at a Leader Identification Task

The Limits of Reactive Shepherding Approaches for Swarm Guidance

Tracking Footprints in a Swarm: Information-Theoretic and Spatial Centre of Influence Measures

Contact Info

Product

Resources

About