ASSISIbf
Animal‌ and‌ robot‌ Societies‌ ‌Self-organise‌
and‌ Integrate‌ by‌ Social‌ Interaction‌ (bees‌ and‌ fish)‌

ASSISIbf Summer School

Graz, Austria, August 29 – 31, 2017

DEADLINE EXTENSION – July 26, 2017

ENGINEERING AND EVOLUTION OF BIO-HYBRID SOCIETIES

Call for applications (PDF)

The goal of this summer school is to teach state-of-the-art methods by going beyond bio-inspired systems, which are focused on developing technology, to bio-hybrid systems, where technology resides in symbiosis with living systems. These bio-hybrid systems can make the best use of the properties of both components: biological and technological. For this, both systems and their interactions need to be understood and modelled in a more detailed way than it would be necessary for classical bio-inspired systems. During this summer school, examples will come from hybrid systems involving robots and bees, fish and plants. The school will consist of four main parts: preparation, lectures, practicals, and a reporting for students who would like to submit the result of their practical work for evaluation.

Lectures

  • From Honey bees to bio-hybrid robot swarms – Thomas Schmickl (University of Graz – Artificial Life Lab)

In this talk I present basic principles of swarm-intelligence and self-regulation in honey bees and how they can be derived from natural observations. I show how these findings can be translated into bio-inspired algorithms first (to be used on robots) and how robots then can be used to be re-integrated into honey bee societies, turning it into a bio-hybrid society.

  • The honey bee – a short introduction – Martina Szopek (University of Graz – Artificial Life Lab)

In this talk I will introduce our model organism, the western honey bee (Apis mellifera), to the participants. I will give a general overview of honey bee biology and behaviour and further focus on the bees’ perception of different physical stimuli that will be used during the practicals involving bees.

  • Lessons on using animals as fitness providers for EA – Pedro Mariano (University of Lisbon – BioISI) & Ziad Salem (University of Graz – Artificial Life Lab)

In this talk we will present evolution of a vibration pattern as a stopping signal for bees. From this case study we extracted several lessons on using animals as fitness providers for an evolutionary algorithm. This talk will describe the experiment, the identified challenges and initiatives to circumvent them.  A generalisation of these results is suggested.

  • Modelling sizes, links and characteristic times of distal societies – Luís Correia & Rob Mills (University of Lisbon – BioISI)

This talk will analyse three features that play a major role in enabling bio-hybrid societies:  information links among societies, characteristic reaction times of the individuals, and sheer number of individuals in each society.  It will be shown that any of these features has limits beyond which a bio-hybrid society is impracticable.

  • Lessons learned from fish-robot experiments – Frank Bonnet (École Polytechnique Fédérale de Lausanne – LSRO & Université Paris Diderot – LIED)

A brief overview of the challenges and some solutions for them, as they were learned in the ASSISIbf project. For example: How to design a robot for experimentally researching fish swarm behaviour? How to design a clever experiment to validate the acceptance of this robot by the fish society? Further details to be announced.

 

Practicals

  • Engineering and design choices in the evolution of robot-bee interaction (University of Lisbon – BioISI & University of Graz – Artificial Life Lab)
  • Adaptation of robot controllers in experiments involving mixed groups of fish and robots (École Polytechnique Fédérale de Lausanne – Université Paris Diderot – LIED)
  • Extract social interaction patterns from a minimal society (University of Graz – Artificial Life Lab)
  • Investigating whether or not fish can count (University of Graz – Artificial Life Lab)
  • Enki simulation of distal societies – links, CT, size (University of Lisbon – BioISI)
  • Counting the bees – competition (test algorithms for swarm size estimation prepared by participants) (University of Zagreb – LARICS)

 

Involvement of participants

We expect from each participant the following effort:

-       Preparation work: 6 paper reading (10 hours)

-       Talks (5 hours)

-       Practicals (16 hours)

-       Analysis and reporting (9 hours)

Total: 40 hours

 

Practical issues

Lectures and practicals will take place at the University of Graz, Artificial Life Lab (http://zool33.uni-graz.at/artlife/ ) and the Institute of Zoology (https://zoologie.uni-graz.at/en/zoology/ ). The summer school will provide coffee breaks. Travel, accommodation and meals are at the expense of the participants. Accepted participants will have to pay a fee of 50,- EUR in advance.

 

Application

Anyone interested in participating should submit a CV and a short motivation letter to martina.szopek@uni-graz.at  by July 23  26, 2017. Applicants will be notified of their acceptance by July 31, 2017. The number of participants will be limited to the available places for the practicals, students will be given priority admittance.