[The Hive Mind in Southampton]

Robots Multi-agent systems Games AI Added on 16/03/2015

The Hive Mind exhibition

I had the privilege to present The Hive Mind, my new sci-fi puzzle game about insect-inspired construction with robots, at the University of Southampton Science and Engineering day last weekend. A lot of people came to play it and it was great to watch how they interacted with it - kids were often better at the game than their parents!

Termites, ants, wasps and other social insects build their nests to be often 100s or even 1000s of times larger than a single individual. The insects have no overall plan, they have no idea what other particular individuals are doing and in case of termites and some ant species, they are blind and never fully know what they created. Instead, they use stigmergy, a principle where a change in the environment, for example the presence of a built wall, encourages and shapes future actions of individuals around. The insects use simple rules that tell them what to build when they sense a certain structure around them, or, in case of termites, when they sense certain amounts of a chemical substances, called pheromones, around.

Screenshot of the sanbox modeOne day, we could use the same principles of decentralised construction in a swarm of builder robots. Why bother? If we sent only a single robot to a remote location, for example to another planet, and the robot broke down, the whole mission would fail. But instead, we could send simple robots with insect-like behaviour to cooperate and build together, making the whole team more robust to failures of single individuals. However, designing the control algorithms for such robot systems is complicated and we don’t fully understand how to do it reliably yet.

In The Hive Mind game, the player takes on the role of a Construction Coordinator who was sent to an alien planet and designs the rules for builder robots. The game offers various challenges, where a certain structure needs to be built by the robots. The higher the challenge level, the more decentralised and complicated the rules need to be. Everyone on the Science and Engineering day started with a training tutorial that teaches the basic principles of how to play. The adults were usually interested in the science behind the game and it was great to see their fascination as we explained how insect construction works. The kids of course did not care about the science and just wanted to build - some creations in the sandbox game mode were really nice!

The Hive Mind exhibitionThe event lasted for 6 hours and there were so many people to talk to. Luckily, I had help from my supervisor, Prof. Seth Bullock, who also cooperated on game development, and Nick Hill, another PhD student who does research into termite construction. Seth and me now have a lot of ideas on how to improve the game and where to take it next. I think playing it is great not only for training your logic, but also for understanding the new paradigm in programming and engineering - decentralisation. As the systems we design are becoming more and more complicated, we need new ways of thinking about how the components that make them should interact with each other with minimal human intervention. The evolution has shaped many species to solve the problem of decentralised control and we can learn a lot by studying the nature.

I hope that one day we can really use robotic swarms on remote planets, under water or even in our bodies to create and repair structures or to gather minerals, water and other things necessary for human survival. NASA is already taking steps towards making it happen. I am contributing in my own small way by looking at how we can learn principles of decentralised control from insects and apply them on robots. And in The Hive Mind, you can contribute too by finding new ways for robots to build interesting things.

You can read even more about The Hive Mind and the event in the EPSRC CDT blog by Hans Fangohr.



{Please enable JavaScript in order to post comments}

Designing robot swarms

In software engineering, a design pattern associates a particular class of known problem with a particular class of effective solution. Analogously, swarm robot engineers would benefit from design patterns that each associate specific robot control schemes with desired collective performance. In this project, we characterise such design patterns for robot swarms in the context of collective foraging and task allocation.

How will artificial life impact the future?

In 2013, I was a part of the TRUCE workshop at the Alife conference. The workshop brought together scientists and creative writers in order to create a cool book full of stories about A-life (artificial life) and artificial intelligence and about how it will impact our society in the future. As I am very interested in swarm robotics, sci-fi games and movies and generally how the future will look like when robots run around and are part of our everyday lives, I jumped at the opportunity to cooperate on the project.

Controlling Ant-Based Construction

Stigmergy allows insect colonies to collectively build structures that no single individual is fully aware of. Since relatively minimal sensory and reasoning capabilities are required of the agents, such building activity could be utilised by robotic swarms if we could learn how to control the shape of the final structures.

Foraging Strategies in Nature and Their Application to Swarm Robotics

While foraging is a task often experimented with in swarm robotics, it is often the case that foraging strategies inspired by nature are chosen without careful consideration. Foraging strategies including solitary foraging, behavioural matching, stigmergy, signaling to guide others and coordinated and cooperative hunting are identified and their implementation costs in robots, as well as their suitability for different scenarios is discussed.

The Centralised Mindset and Complexity Science

Humans tend to explain decentralised phenomena as being caused by a single entity. This way of thinking is often referred to as 'the centralised mindset'. Several authors propose that using programming environments where creation of decentralised agent-based systems is easy...

pyCreeper

The main purpose of pyCreeper is to wrap tens of lines of python code, required to produce graphs that look good for a publication, into functions. It takes away your need to understand various quirks of matplotlib and gives you back ready-to-use and well-documented code.