Artificial Cell Systems Based in Gene Expression Protein Effects - Advancing Artificial Intelligence Through Biological Process Applications - page 162

FUTURE RESEARCH DIRECTIONS

Finally, remarks that even the model showed
here have been used for the development of
forms; it could be used to develop other task
like the processing information. This point can
be development using the genetic regulatory
networks theory so rules can be defined as the
combination of different genes. This theory has a
Boolean algebra like structure as it could be seen
in (Kauffman, 1969).

On the last years the artificial embryogeny
has had a very quickly development. As it was
mentioned, models which are sorted as artificial
embryogeny kind could be framed into the gram-
matical approach or into the chemical approach.
On the last years, a lot of grammatical approach
models have been adapted to generate artificial
neural networks. This work is an interesting
point of investigation, because it has application
on evolvable artificial neural networks, such as
archive an evolvable design of a network for a
particular problem.
On the chemical approach models, as the one
presented in this paper, could also include new
characteristics such as the displacement of cells
around their environment, or a specialisation
operator that blocks pieces of DNA during the
expression of its descendants, as happens in the
natural model. These incorporations may induce
new behaviours that make it applicable to new
problems.
One of the questions is to develop models which
could combine both approaches. The result will
be a hybrid model which will has characteristics
from both approaches.
Explore other possible applications of the
artificial embryogeny models could be an inter-
esting works in areas like the evolutionary robot
controllers design, the evolutionary hardware,
design of generative encoding for the construc-
tion of artificial organisms in simulated physical
environments, etc.
Another problem that could be easy detected is
to find a strategy search to work with the model.
Both approaches, chemical and grammatical,
have to search into enormous search spaces.
Many models have failed into their applicabil-
ity because it was difficult to search the correct
configuration for the rules of the system. Find
a general way to develop that search may be an
interesting work.

REFERENCES

Bentley, P.J. (2002) Digital Biology . Simon and
Schuster, New York.

Conway J.H. (1971) Regular Algebra and Finite
Machines . Chapman and Hall, Ltd., London.

Dellaert F., Beer R.D. (1996) A Developmental
Model for the Evolution of Complete Autonomous
Agent In From animals to animats: Proceedings
of the Forth International Conference on Simula-
tion of Adaptive Behaviour, Massachusetts, 9-13
September 1996, pp. 394-401, MIT Press.

Eggenberger P. (1996) Cell Interactions as a
Control Tool of Developmental Processes for
Evolutionary Robotics In From animals to
animats: Proceedings of the Forth International
Conference on Simulation of Adaptive Behaviour,
Massachusetts, 9-13 September 1996, pp. 440-
448, MIT Press.

Endo K., Maeno T., Kitano H. (2003) Co-evolu-
tion of morphology and walking pattern of biped
humanoid robot using evolutionary computa-
tion -designing the real robot . ICRA 2003: pp.
1362-1367

Fernandez-Blanco E., Dorado J., Rabuñal J.R.,
Gestal M., Pedreira N.(2007) A New Evolutionary
Computation Technique for 2D Morphogenesis
and Information Processing . WSEAS Transac-
tions on Information Science & Applications Vol-
ume 4 pp. 600-607. April 2007. WSEAS Press.