Self-Organized Collective Decision-Making in a 100-Robot Swarm

Authors: Gabriele Valentini, Heiko Hamann, Marco Dorigo

AAAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental We study the effectiveness and robustness of the proposed strategy using a swarm of 100 Kilobots and we focus on the impact of the neighborhood size over the dynamics of the system.
Researcher Affiliation Academia Gabriele Valentini IRIDIA, Universit e Libre de Bruxelles 50 Av. F. Roosevelt, Brussels, Belgium +32 2 650 2712, gvalenti@ulb.ac.be, http://iridia.ulb.ac.be/ gvalentini/ Heiko Hamann Department of Computer Science University of Paderborn Paderborn, Germany heiko.hamann@uni-paderborn.de Marco Dorigo IRIDIA Universit e Libre de Bruxelles Brussels, Belgium mdorigo@ulb.ac.be
Pseudocode No The paper describes the robot's control algorithm as a "finite-state machine of 4 states (Fig. 1a)", and describes its steps in text, but it does not provide formal pseudocode or an algorithm block.
Open Source Code No The paper does not contain any statement about releasing source code or provide a link to a code repository.
Open Datasets No The paper describes a custom experimental setup with 100 Kilobots in a rectangular arena where sites A and B are emulated with infrared beacons. This is an custom experimental environment, not a publicly available dataset with concrete access information.
Dataset Splits No The paper does not specify explicit training, validation, or test dataset splits. It describes experimental runs (10 runs of 90 min) and collects data dynamically from the robot swarm.
Hardware Specification Yes The Kilobot is a low-cost, 3.3 cm robot with stick-slip motion (1 cm/s forward, π/4 rad/s turn in place), one ambient light sensor, and infrared communication capabilities (3-byte messages in a range up to 10-20 cm). We place N = 100 robots in a rectangular arena of 100 x 190 cm2 (Fig. 1b).
Software Dependencies No The paper describes the distributed algorithm and the robot's finite-state machine logic, but it does not list any specific software dependencies with version numbers (e.g., programming languages, libraries, operating systems, or specific simulation software).
Experiment Setup Yes We place N = 100 robots in a rectangular arena of 100 x 190 cm2 (Fig. 1b). We consider a scenario where site A is twice as good compared to site B (ρA = 1 and ρB = 0.5). We perform two series of experiments (10 runs of 90 min for each series) where we vary the maximum number Nmax of opinion messages that a robot is allowed to receive before applying the majority rule. For convenience, we refer to the maximum size Gmax = Nmax + 1 of the group of opinions used in the majority rule which includes the robot current opinion (Gmax {5, 25}).