Course Title:Bio-Inspired RoboticsMohammad Iqbal

(Thanks to Masoud Asadpour)

Lecture: Control of Collective Behavior

In the name of God

Some of the slides have been taken from the presentations of J.Halloy et al, ULB on LEURRE project

Smart Collars and Virtual Fences

http://groups.csail.mit.edu/drl/wiki/images/c/ca/bcpr-cows.pdf

Centralized Control of collective behavior

Distributed Control with Informed Agents

In many collective behaviors, only a fewindividuals among a group that, e.g. forage or travel, have pertinent information, such as knowledge about the location of a food source, or of a migration route. Couzin et al (Nature 433, 2005) showed that the larger the group, the smaller the proportion of informed individuals needed to guide the group, and that only a very small proportion of informed individuals is required to achieve great accuracy. Potential applications in human society: Education, Dialog between cultures, Government, Stock market, Fashions.

LEURRE Project: Realization of Collective Behavior Control

Cockroaches tend to aggregate in darkplaces. Can we change their behavior and force them in some way to select bright places?Solution: Use some robot which seem like cockroaches (in their eyes!)

Polyethylene ring (diameter: 100cm, height: 20cm, thickness: 1cm)

Camera: Panasonic WV-BP330

Plexiglas shelters covered by red filters

©Philippe PLAILLY / EURELIOS

Experimental protocol

0.0060.030Cockroach

Dark shelterLight shelter

Probability of leaving+/-

+Shelters

Collective selection

>

Central role of the number of neighbors on cockroach behavior: P of leaving shelter decreases with increasing number of neighbors.

Influence of shelter darkness on P of leaving

Modulation of cockroach’s behaviour

0

0.1

0.2

0.3

0.4

0.5

0.6

0-0.33 0.34-0.66 0.67-1

fraction of sheltered cockroaches under the shelter 1

Frac

tion

of e

xper

imen

ts

0

0.2

0.4

0.6

0.8

1

Winner shelter Loser shelterFrac

tion

of c

ockr

oach

es u

nder

she

lter

Selection between 2 identical shelters

10 cockroaches/light shelters10 cockroaches/dark shelters

30 cockroaches/dark shelters

Symmetry

Collective selection of only ONE resting site

Collective selection of the darkest shelter

Discrimination increases with group size 0

0.2

0.4

0.6

0.8

0-0.33 0.34-0.66 0.67-1

fraction of sheltered cockroaches under the dark shelter

Frac

tion

of e

xper

imen

ts10 cockroaches (n=30)

12 cockroaches (n=30)

30 cockroaches (n=25)

Collective selection between 2 different shelters

Collective decision making in cockroach group

bifurcations leading to multiple steady states

Mixed Society

Mixed groups are as selective as pure cockroaches groups

0

0.2

0.4

0.6

0.8

1

0-19 20-39 40-59 60-79 80-100Percentage of individuals under the winner shelter

Frac

tion

of e

xper

imen

ts

without insbot (n=13)

with insbot (n=21)

Collective choice in mixed-society Homogeneous shelters

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0-19 20-39 40-59 60-79 80-100

Percentage of individuals under the dark shelter (%)

Frac

tion

of e

xper

imen

ts w ithout insbot (n=29)

w ith insbot (n=14)

Collective choice in mixed-society Heterogeneous shelters

The proportion of experiments choosing either the dark or the light shelter is same in pure

cockroaches and mixed societies

Insects prefer darkRobots prefer light

Change in collective choice

Preference of the robots for dark and light shelters is swapped. So they prefer the light shelter instead of the dark oneThe robots are able to induce a change in the global pattern by reversing the collective shelter preference.

0.59

0.16

0.320.38

0.03

0.59

0

0.2

0.4

0.6

0.8

0 - 0.33 0.34 - 0.66 0.67 - 1

Fraction of individuals under the dark shelter

Frac

tion

of e

xper

imen

ts

bright preference dark preference

0

5

10

15

20

25

30

35

2 cockroaches marked unmarked

Mea

n tim

e of

con

tact

(s)

A

Importance of chemical marking

10 xMarked or unmarked1 x

SummaryCockroaches perform group choice that is a form of self-organized collective decision. It emerges form the local interactions between individuals.Both machines and insects are capable, independently of each other, to perform such collective decision.The robots are accepted by the cockroaches groups and actively take part in the collective choice.Most of the time, they gather with the cockroaches under the same shelter.When the robots are programmed to have an opposite preference compared to insects, they are able to induce a change in the global pattern by reversing the collective shelter preference.The mixed group of robots and insects gather in the less preferred shelter by the insects.

These experimental results demonstrate the existence of shared and controlled collective choice between machines and animals.