The Behavior of a Swarm of Simple Robots Expressing Collective Intelligence
CSEF · 2016 Electronics & Electromagnetics Third Award
Overview
Objectives/Goals The objective of this study was to design a swarm of tiny robots modeled off of ants that were able to accomplish complex tasks, despite each individual being limited in its abilities. I tested out the existence of the four main characteristics of swarm robotics in these robots: collaboration, scalability, robustness, and most of all, low cost. At the end I made the robots do a final task, which was a simulation of an ant colony searching for food in order to see whether my robots actually qualified as swarm robots with collective intelligence. Methods/Materials Attiyn85, LED, Phototransistor, DC motors, MOSFET, Li-ion Coin Battery. I created a palm-sized robot that was equipped with only a phototransistor, an LED, and two DC motors. I designed several different kinds of simulations such as dispersement and synchronization experiments. I tested whether the tiny robots showed signs of collective intelligence depending on whether they could complete the given task. Results Even though I did not directly implement any kind of work distribution program, the robots were able to evenly distribute the work of scouting for an area among themselves. The graph of task efficiency showed a linear slope, indicating these tiny robots showed decent efficiency for its simplicity. The test for robustness also showed robots being able to pass on information even if several robots malfunctioned in between. Each robot costed around $15 to make, and took about a whole day to build, achieving low cost. Conclusions/Discussion The tiny robots, having only two input sensors that perceived its surroundings, were capable of having communications with one another, completing the final simulation. The robots were capable of showing all four characteristics of swarm robotics, proving that my project was successful. Although these experiments were done under ideal conditions, it proves the validity of this concept of using simple robots to have them accomplish a larger task as a whole, eventually leading to future applications such as rescue robots during natural disasters.
Summary statement
I designed a functional swarm of tiny robots using only the bare minimum parts, proving the possibilities of this type of swarm robotic system.
Help received
From the prototyping and the actual production, the tiny robots were made completely by myself, along as conducting the experiments. I received help from my father in software development of these robots, who taught me how to program.
Awards (1)
Competition history
- CSEF 2016
Resources
Related projects
CSEF · 2026
Ant Colony Simulation: The Power of Swarm Intelligence
CWSF · 2026
Emergent Collective Intelligence in Swarm Robotics via Stigmergic Multi-Agent Reinforcement Learning
ISEF · 2015
The Implantation of the Phototaxic Behavior of Physarum polycephalum in a Decentralized Robotics System
CSEF · 2015
Little Insect Project (Inspired by DARPA LittleDog Project)
CSEF · 2016
An Application of Artificial Neural Networks to Decentralized Cooperative Navigation
CSEF · 2017
Decoding the Artificial Intelligence behind Tactile and Photocell Navigation
CSEF · 2014
Replicating Biological Walking Patterns in Robotic Systems
ISEF · 2024
Effective Robotic Swarm Controller Applied to Autonomous Swarm Assembly of Modular Flexible Production Lines
Closest projects by meaning, across every fair and year in the corpus.
Browse more like this
Source: California Science & Engineering Fair public projects