Path formation in a robot swarm

Abstract

We present two swarm intelligence control mechanisms used for distributed robot path formation. In the first, the robots form linear chains. We study three variants of robot chains, which vary in the degree of motion allowed to the chain structure. The second mechanism is called vectorfield. In this case, the robots form a pattern that globally indicates the direction towards a goal or home location.We test each controller on a task that consists in forming a path between two objects which an individual robot cannot perceive simultaneously. Our simulation experiments show promising results. All the controllers are able to form paths in complex obstacle environments and exhibit very good scalability, robustness, and fault tolerance characteristics. Additionally, we observe that chains perform better for small robot group sizes, while vectorfield performs better for large groups.

F-Race and Iterated F-Race: An Overview

“…Most applications focus on configuring SLS methods for combinatorial optimization problems [9,2,25,27,31,40,41]. However, also other applications have been considered including the tuning of algorithms for training neural networks [15,47], or the tuning of parameters of a control system for simple robots [37,38].…”

A review of swarm robotics tasks

“…Pheromone communication has been extensively studied for the foraging task, where it is used to dynamically build optimized paths between the nest and rich food sources [70,71,73]; a mechanism analogous to the use of pheromone is put in place when robots exchange local messages whose content provides an indication of the current position of communicating robots with respect to the path being formed [112]. In probabilistic methods, a chain of robots is created as a sequence of stochastic events determined by robots continuously joining and leaving the chain; this dynamic process allows robots to explore new areas in the environment until both locations of interest are found and an optimized path between them is formed [113]. In evolutionary methods, robot behavior is evolved according to a fitness function that measures the quality of a path established between two locations: for example, in [114] robots are controlled thorough a simple neural network whose parameters are evolved with a genetic algorithm, and the fitness function rewards robots based on how many times they travel from one target location to the other.…”

Group abstraction for assisted navigation of social activities in intelligent environments

“…Typically in swarm robotics, success is achieved when some number of robots reach the goal or find a path [13,29,32,31,42,30]. However, for ACANTO, an activity is only considered completely successful if all the social group members complete the activity, and do so coherently.…”