Evolving Aggregation Behaviors in Multi-Robot Systems with Binary Sensors

Abstract

This paper investigates a non-traditional sensing trade-off in swarm robotics: one in which each robot has a relatively long sensing range, but processes a minimal amount of information. Aggregation is used as a case study, where randomly-placed robots are required to meet at a common location without using environmental cues. The binary sensor used only lets a robot know whether or not there is another robot in its direct line of sight. Simulation results with both a memoryless controller (reactive) and a controller with memory (recurrent) prove that this sensor is enough to achieve error-free aggregation, as long as a sufficient sensing range is provided. The recurrent controller gave better results in simulation, and a post-evaluation with it shows that it is able to aggregate at least 1000 robots into a single cluster consistently. Simulation results also show that, with the recurrent controller, false negative noise on the sensor can speed up the aggregation process. The system has been implemented on 20 physical e-puck robots, and systematic experiments have been performed with both controllers: on average, 86-89% of the robots aggregated into a single cluster within 10 minutes.

Embodied Evolution in Collective Robotics: A Review

“…Recent works in reinforcement learning have developed theoretical tools to break down complexity by operating a move from considering many agents to a collection of single agents, each of which being optimized separately (Dibangoye et al, 2015), leading to theoretically well-founded contributions, but with limited practical validation involving very few robots and simple tasks . Lacking theoretical foundations, but instead based on the experimental validation, swarm robotics controllers have been developed with black-box optimization methods ranging from brute-force optimization using a simplified (hence tractable) representation of a problem (Werfel et al, 2014) and evolutionary robotics (Hauert et al, 2008;Trianni et al, 2008;Gauci et al, 2012;Silva et al, 2016).…”

A distance weighted-based approach for self-organized aggregation in robot swarms

“…In a similar setup, Soysal et al [18] investigated the effects of a number of parameters, such as the robots number, the size of arena, and run time. In another study, Melvin et al [19] proposed two algorithms -a reactive controller with no memory and a recurrent controller with memory-to study aggregation in a swarm of e-puck robots. The algorithms are based on classical evolutionary programming technique, and use simple binary sensor that is proven to be enough to achieve error-free aggregation since a sufficient sensing range is provided.…”

Distributed Particle Swarm Optimization - particle allocation and neighborhood topologies for the learning of cooperative robotic behaviors

“…Some researchers have shown that it is possible to use learning techniques to generate cooperative behaviors [2], [3], [12], [13]. Mataric [12] and Parker [13] addressed the topic of learning in multi-robot teams using a small number of parameters per robot, as opposed to the large search space addressed in this paper.…”

“…It should be mentioned that the task as implemented in this article is harder than those from other works in that the robots are not physically connected to each other [2], they are required not only to aggregate but also move together [3], and there is no environmental template or goal to guide their movement [1]. In the case of [2] and [3] learning has been done only in a centralized manner, using homogeneous controllers and a global performance metric.…”

“…In the case of [2] and [3] learning has been done only in a centralized manner, using homogeneous controllers and a global performance metric.…”