Adaptive collective decision-making in limited robot swarms without communication

Abstract

In this work we investigate spatial collective decision-making in a swarm of microrobots, inspired by the thermotactic aggregation behavior of honeybees. The sensing and navigation capabilities of these robots are intentionally limited; no digital sensor data processing and no direct communication are allowed. In this way, we can approximate the features of smaller mesoscopic-scale systems and demonstrate that even such a limited swarm is nonetheless able to exhibit simple forms of intelligent and adaptive collective behavior

A review of swarm robotics tasks

“…Taking inspiration from social insects, a number of probabilistic robot control algorithms achieve aggregation using a finite state machine characterized by two basic states: walk and wait [20,21,22,23,24,25,26,27,28,29,30,31,32]. At 9 each state corresponds a specific robot behavior; the different states in use (and the behavior implemented in each state) vary between algorithms, but many similarities can be found in various aggregation studies.…”

“…In other studies [21,23,50], two distinct light sources are put in the environment, and the aggregation dynamics varies with the relative size and intensity of the light spots. A particular case of the scenario with two light sources in the arena is when the two sources are identical.…”

“…Once a suitable definition of aggregates is adopted, performance metrics can be calculated as the ratio of the number of robots forming the largest aggregate to the total number of robots [45,31,44], or as the average aggregate size [33,52]; a more in-depth analysis of the aggregation dynamics can be conducted by observing the distribution of robots belonging to aggregates of different size [35]. In tasks where the objective is to aggregate robots in a given area, a common metric is given by the number of robots located in the target area [26,53,32,28,21,48] or within a certain distance from a defined aggregation spot [22,50].…”

“…In algorithms where the aggregation process is influenced by local characteristics of the environment, mathematical models must take into account such characteristics, and the collective dynamics can be expressed either with the number of robots inside the areas designated for aggregation [21], or with the density of robots expressed as a function of spatial coordinates [53,54]. In [20] an experimental setup to simulate honeybee aggregation is described, and the probability of forming aggregates of a certain size in a given area is put 22 in relation with the luminance of the area and other parameters such as the density of robots in the arena, their speed and their sensing range.…”

Decentralized self-selection of swarm trajectories: from dynamical systems theory to robotic implementation

“…Weak emergence is here understood in the S. and O. Kernbach's interpretation (Kernbach 2013;Kernbach et al 2013) (i.e., macroscopic swarm behavior due to interactions between agents but having its origin in the model that describes the system's behavior). Our model has a tractable complexity, enabling us to explicitly describe the resulting emergent collective behavior, which is controllable.…”