Optimal network topology for responsive collective behavior

Abstract

Animals, humans, and multi-robot systems operate in dynamic environments, where the ability to respond to changing circumstances is paramount. An effective collective response requires suitable information transfer among agents and thus critically depends on the interaction network. To investigate the influence of the network topology on collective response, we consider an archetypal model of distributed decision-making and study the capacity of the system to follow a driving signal for varying topologies and system sizes. Experiments with a swarm of robots reveal a nontrivial relationship between frequency of the driving signal and optimal network topology. The emergent collective response to slow-changing perturbations increases with the degree of the interaction network, but the opposite is true for the response to fast-changing ones. These results have far-reaching implications for the design and understanding of distributed systems: a dynamic rewiring of the interaction network is essential to effective collective operations at different time scales.

Coherent collective behaviour emerging from decentralised balancing of social feedback and noise

“…For instance, in (Huepe et al 2011;Chen et al 2016) the authors considered dynamically changing building blocks of adaptive networks and analytically derived their influence on the swarm decision. Furthermore, network-theoretic concepts were applied to analyse the impact of the number of interactions on flocking dynamics and collective response to an oscillating signal (Shang and Bouffanais 2014;Mateo et al 2017Mateo et al , 2019. Similarly, Khaluf et al (2017aKhaluf et al ( , 2018 highlighted the role of different interaction models in enabling the system to restore a specific level of social feedback necessary for convergence to a collective decision.…”

Integrated Robotic and Network Simulation Method

“…The main idea is to deploy the solution in hardware and see how all the algorithms work together. This approach led to positive results as shown in [27] for swarm response and in [28] for topology dynamics, but it requires hardware implementation and can be exhausting when several testing cases are required. Our proposed approach differs from [26] because ours is simulation-based, addressing the case when there is a need of exhaustive simulations or when the hardware testbed is unavailable.…”

A Sensor-Network-Supported Mobile-Agent-Search Strategy for Wilderness Rescue

“…Outside of search and rescue applications, other works have focused on the design of multiagent systems operating in dynamic environments for effective operation at different time scales. In reference [44], results have shown that dynamic changes to a multi-agent network are critical to Earlier solutions to this problem have focused on using solely mobile agents [12][13][14][15][16][17][18][19][20][21][22][23][24][25][26]. For example, in reference [22], a formation-based search method is presented.…”

“…Outside of search and rescue applications, other works have focused on the design of multi-agent systems operating in dynamic environments for effective operation at different time scales. In reference [44], results have shown that dynamic changes to a multi-agent network are critical to efficient collective performance at various time scales. Other works have investigated the group controllability of multi-agent systems in dynamic environments for real-time applications [45,46].…”