Co-evolution of Morphology and Control of Soft-bodied Multicellular Animats - supplementary materials

This page contains videos that supplement figures for the GECCO 2012 (Genetic Algorithms and Genetic Programming Conference) paper by Michał Joachimczak and Borys Wróbel.

If you have any questions or comments, please don't hesitate to contact me (Michal): mjoach ( at ) gmail . com

The paper can be downloaded from the link at the bottom of the page.

Contents:

  1. Abstract
  2. Figure 4 videos
  3. Figure 5 videos
  4. Figure 6 videos
  5. Reference/BibTeX
  6. Related work on the system

1. Paper abstract

We present a platform that allows for co-evolution of development and motion control of soft-bodied, multicellular animats in a 2-dimensional fluid-like environment. Artificial gene regulatory networks (GRNs) with real-valued expression levels control cell division and differentiation in multi cellular embryos. Embryos develop in a simulated physics environment and are converted into animat structures by connecting neighboring cells with elastic springs. The springs connecting outer cells form the external envelope which is subject to fluid drag. Both the developmental program and motion control are encoded indirectly in a single linear genome, which consists of regulatory regions and regions that code for regulatory products (some of which act as morphogens). We applied a genetic algorithm to co-evolve morphology and control using a fitness measure whose value depends on distance traveled during the evaluation phase. We obtained various emergent morphologies and types of locomotion, some of them showing the use of appendages



2. Figure 4 - locomotion of two undulating individuals (high drag environment)

This and following figures demonstrates a sample of obtained morphologies and locomotion types in the system. Note that the individuals were evolved only for their ability to move in a fluid. Neither their morphology nor the type of controller were externally enforced. Each individual develops from a single cell into its final morphology and the behavior of each cell during development is controlled by the activity of its gene regulatory network encoded in linear genome. The locomotion of individual is controlled by oscillators, whose frequency, phase shift and amplitude is set by the concentrations of selected transcription factors.



Fig. 4a Undulating individual
Fig. 4b Another example of undulating individual



3. Figure 5 - Individuals propelling themselves through asymmetric contractions (moderate drag environment)

This and following figures demonstrates a sample of obtained morphologies and locomotion types in the system. Note that the individuals were evolved only for their ability to move in a fluid. Neither their morphology nor the type of controller were externally enforced. Each individual develops from a single cell into its final morphology and the behavior of each cell during development is controlled by the activity of its gene regulatory network encoded in linear genome. The locomotion of individual is controlled by oscillators, whose frequency, phase shift and amplitude is set by the concentrations of selected transcription factors.



Fig. 5a Individual propelling itself through asymmetric contractions
Fig. 5b Individual propelling itself through asymmetric contractions



4. Figure 6 - individuals propelling themselves with the use of symmetric protrusions (low fluid drag environment)

This and following figures demonstrates a sample of obtained morphologies and locomotion types in the system. Note that the individuals were evolved only for their ability to move in a fluid. Neither their morphology nor the type of controller were externally enforced. Each individual develops from a single cell into its final morphology and the behavior of each cell during development is controlled by the activity of its gene regulatory network encoded in linear genome. The locomotion of individual is controlled by oscillators, whose frequency, phase shift and amplitude is set by the concentrations of selected transcription factors.



Fig. 6a Individual with symmetric appendages (32 cells)
Fig. 6b Individual with symmetric appendages, cell limit increased to 64 cells



5. Reference/Bibtex

Joachimczak, M. and Wróbel, B. (2012). Co-evolution of morphology and control of soft-bodied multicellular animats. In Soule, T. et al., editors, Proceedings of the Fourteenth International Conference on Genetic and Evolutionary Computation, GECCO '12, pages 561-568, New York, NY, USA. ACM.
 PDF (ACM) /  PDF (Preprint) /  BibTeX /  RIS /   CiteULike


6. Related work on the system

Have a look at a related paper, in which the cells can actively control their behavior with GRN (rather than evolve parameteres for oscillators):

http://www.evosys.org/grnanimats

as well as at our earlier work on the same model of gene regulatory network, applied to 3D development and patterning:

http://www.evosys.org

If you can't access the pdf or it's not yet available, just e-mail me: mjoach (at) gmail . com


Last modified: May 2013, by Michał Joachimczak