Subtask 1 - Morphogenesis

Report Jan 16th - 16 Jan 2012 08:31

During the meeting on Jan. 11th, it was decided that one explicit and one implicit representation will be used for the demo. All representations will be presented, evaluated and discussed in the report.

Explicit representation

Wenguo's explicit representation is chosen as the most advanced.

implicit representations:

Virtual Embriogeny (Graz)

Ported on Robot3D (on-going), and some measurements are available:

  • Size of the genome, from 50 to 500, (depending on the complexity, not size, of organism)
  • Viability, 5 out of 1,000 random genomes
  • Needs about 150 generations of 100 individuals (10 ** 4) to reach size 10.
  • Controllability: no information yet.
  • Diversity of shapes: some examples of shape are attached to this wiki

GRN (Ghent)

GRN is an explicit representation

Cellular Automaton (Paris)

Ported on Robot3D (magic docking):

  • Size of the genome: 30
  • Viability, circa 10%
  • Max size: about 5,000 fitness evaluations to reach size 10. Improving.
  • Curves reporting number of different shapes vs number of offspring (resp. number of viable offspring), averaged out of 10 runs, are available on the slides. attached.
  • Limited controllability: small cross is OK; big cross, not yet.

Summary about implicit representations

Not all information is available yet. We have 2 possibilities:

  • Guzs decides (pros of VE: more fancy; pros of CA: simpler)
  • See if we can have all information on Jan. 19th and decide face to face.

Tags: | Comments: 0

6 - 15 Jan 2012 22:18

Tags: | Comments: 0

Report from the decision meeting (11) Jan - 11 Jan 2012 15:32

Answer from Graz (Skype meeting, Jan. 13th and mail 15th of January)

  • Size of the genome (Virtual Embryogenesis): between 50 and 500 depending on the shape complexity
  • Viability: 5 genomes viables out of 1,000 random genomes.
  • Mapping is deterministic (except for the fact that there might be no free robots in the environment)
  • Controllability not tested but some examples of shape attached.
  • Max size: goes to 10 after 150 generations on 5 runs (Runs attached). Question, how many individuals per generation ? Let's assume 30.
  • Number of different shapes: not yet.

Answer from Ghent (Skype meeting, Jan. 13th)

  • Size for a shape of size n: 3n
  • A threshold is used; default value considered in the following is .3 for each gene.
  • Out of 1,000 random genomes (with .3 threshold), 80% are viable; 40% are unique, viable shapes
  • Mapping stochastic; ratio phenotype/genotype depends on a threshold; for threshold .3, figure will arrive.
  • Controllability: examples of shape will arrive shortly.
  • Initialization: choose the size; then you have a string; compare each value to the threshold; this makes it an explicit representation.

Next step:

  • Complete the figures for the report
  • Clarify relationship between GRN and AGE.

Summary of the Jan. 11th

Regarding our Morphogenesis task, it was decided that *for the demo* we would consider one explicit representation (Wenguo's is chosen as the most advanced one) and one implicit representation. All representations will be considered
in view of the report and the paper.

There are 3 implicit representations at the moment: Graz's (Virtual Embryogeny), Paris' (Cellular Automaton) and Ghent's (Yao's GRN).

The decision was not taken on the spot out of lack of information: some measurements were missing on GRN and VE.

These measurements are the following (nothing fancy; the usual tables and figures found in every experimental section of every paper in evolutionary computation):

  • what is the size (number of symbols, integers, real-values) of the genotype needed to encode a shape with n robots ? If the genotype to phenotype mapping involves the environment (e.g. the sensors of the robots), consider an average environment/empty environment in Robot3D for the sake of reproducibility.
  • what is the proportion of viable (i.e. less than 10 robots, not 4 robots linked in a square) built from 1,000 genomes drawn uniformly in the search space ?
  • is the mapping (genotype to phenotype) deterministic or stochastic ? In the latter case, how many viable shapes (phenotypes) are built from one genotype on average ?
  • what is the controllability of the representation, i.e.: consider a stair or a cross shape, is there a genotype coding for these shapes ?
  • what is the evolvability of the representation, i.e., take a random population, use what you like as fitness (my recommandation, the size of the organism) and draw
  • ** the max. size of the organism built vs the number of fitness evaluations (average out of 10 runs);

** the number of different shapes (two shapes are equal if they can exactly be super-imposed, egg on egg) vs the number of fitness evaluations (average out of 10 runs);
** the number of different shapes (two shapes are equal if they can exactly be super-imposed while keeping the egg) vs the number of fitness evaluations corresponding to viable shapes (average out of 10 runs);
Tags: | Comments: 0

Week 4 Report - 09 Dec 2011 10:42


Further improvement on explicitly representation of organism
Implemented a simple genetic algorithm on evolving organism shapes;
next: better design of fitness function and parameter tuning
Code is updated in repository


Improving the Genome structure. Installing the Robot3d simulator. Taking into account viability constraints, see below.


Excused for medical reasons


Extension from deterministic to stochastic cellular automata.
Curves (average max. size vs number fitness evaluations; average nb of distinct shapes vs number of fitness evaluations) attached. All result files available at:


a single constraint of morpho viability: dock on a single side at the same time
discussion about which fitness function: size of new shape.


Define primitives: virtual class Egg; primitive Egg.Create_Organism()
Preliminary validation setting: Wenguo's (Player): to collect indicators below.
Use by partners:
* we might need shortcuts: Each free robot knows where the eggs are, and if they recruit; and it goes there directly.


  • Size of genome
  • Properties of the mapping (when implicit)
      • Deterministic / stochastic
      • Ratio phenotypes/genotype
      • Controllability: find genome for a given phenotype
  • Initialization
      • Viability: between 3 and 10 cells
      • Percentage of viable phenotypes from random genotypes
  • Variation operators
      • Number of distinct shapes produced along evolution, from random init, vs number of fitness evaluations.

Tags: | Comments: 0

Week 3 Report - 09 Dec 2011 10:26


A skype meeting is scheduled on Wednesday 30 afternoon - 2:30 pm to discuss and prepare how the task force can decide between the various shape representations/evolution. [Wenguo, Yao-yao, Michele, Ronny hopefully - best wishes of recovery to you Ronny]. Are some others interested ?


(explicit representation, well-formed expression)
has implemented and tested various functions including random initialization and basic variation operators.
His code is available at:

  • Next: documentation.


(virtual embryogeny, AHHS)
has studied the initialization step; initialization from predefined shapes (I) seems more effective at the moment than from scratch. * Ronny was ill this week.


(virtual embryogeny, GRN)
considers a fixed genome size (96kB) which includes various genes (which can be duplicated and modified). Each gene belongs to a status (growing, stable, swarm, aggregate); in growing status, in each robot cell the program searches for the marching gene, which defines the local topology desired for this cell.

  • Next: improving feedback loops to control gene duplication


(virtual embryogeny, probabilistic Cellular Automaton)
has implemented and tested the evolutionary optimization of a cellular automaton defining a shape (pgm MorphoSCA). This week, using deterministic CA, the average curves giving the max. size of the organism (resp. the number of distinct shapes) vs the number of fitness evaluations, with confidence interval, have been obtained (Files Maxsize and Distinct Shapes), attached.

Tags: | Comments: 0

Week 2 Report - 09 Dec 2011 10:21

This is the report for week 2: REPORT_Morpho_B.pdf
Tags: | Comments: 0

Week 1 Report - 09 Dec 2011 10:19

This is the report of Week 1 for Morphogenesis: REPORT_Morpho_A.pdf
Tags: | Comments: 0

Week 26 Dec. 2011

Averages over 10 runs.

Max size of shapes over nb of evaluations: [[]]
Max number of distinct shapes over nb of evaluations: [[]]
Max number of distinct shapes over nb of viable shapes : [[]]

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-NonCommercial-ShareAlike 3.0 License