Evolving Complete L-Systems: Using Genetic Algorithms for the Generation of Realistic Plants

This paper introduces the idea of producing a platform, from which to set into motion the evolution of a virtual plant. The plants are represented and generated using the mathematical formalism of Lindenmayer Systems, with the genome being embodied by the production rules and variables of the L-System, for both the plant and its roots. Evolution is then simulated using Genetic Algorithms, which are driven by a user defined fitness function, in the form of an equation. This is all packaged in a Graphical User Interface (GUI), allowing the user control over the different genetic operators of the genetic algorithm.

[1]  Luc Van Gool,et al.  Procedural modeling of buildings , 2006, SIGGRAPH 2006.

[2]  A. Lindenmayer.  Mathematical models for cellular interactions in development. I. Filaments with one-sided inputs. , 1968, Journal of theoretical biology.

[3]  Kenrick Mock.  Wildwood: the evolution of L-system plants for virtual environments , 1998, 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360).

[4]  Christian Jacob,et al.  Evolution Programs Evolved , 1996, PPSN.

[5]  Radomír Mech,et al.  L-studio/cpfg: A Software System for Modeling Plants , 1999, AGTIVE.

[6]  Przemyslaw Prusinkiewicz,et al.  The Algorithmic Beauty of Plants , 1990, The Virtual Laboratory.

[7]  Radomír Mech,et al.  Visual models of plants interacting with their environment , 1996, SIGGRAPH.

[8]  K. Niklas.  Biophysical limitations on plant form and evolution , 1988 .

[9]  Gabriela Ochoa,et al.  On Genetic Algorithms and Lindenmayer Systems , 1998, PPSN.