Evolving Virtual Creatures
Kenrick E. Drew, Andrew Forsberg, Paton J. Lewis
Abstract
Kenrick E. Drew, Andrew Forsberg, Paton J. Lewis
Abstract
We propose to implement a two-dimensional version of the work described in Karl Sims’s 1994 SIGGRAPH Proceedings paper “Evolving Virtual Creatures.” [SIMSa] We will investigate goaldirected genetic algorithms and the dynamics of complex rigid bodies. Our application will generate animations of virtual creatures exhibiting evolved behaviors. The user will specify the characteristics of the environment in which two-dimensional creatures will evolve. An initial random set of creatures will be created and simulated, and those that are successful at specified behaviors will be selected for breeding. Goal behaviors will include jumping, walking, and swimming. Once the program has bred several generations of creatures, survivors may be saved for future examination.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
We propose to implement a two-dimensional version of the work described in Karl Sims’s 1994 SIGGRAPH Proceedings paper “Evolving Virtual Creatures.” [SIMSa] We will investigate goaldirected genetic algorithms and the dynamics of complex rigid bodies. Our application will generate animations of virtual creatures exhibiting evolved behaviors. The user will specify the characteristics of the environment in which two-dimensional creatures will evolve. An initial random set of creatures will be created and simulated, and those that are successful at specified behaviors will be selected for breeding. Goal behaviors will include jumping, walking, and swimming. Once the program has bred several generations of creatures, survivors may be saved for future examination.
Key concepts: Creatures, Set (abstract data type), Computer science, Human–computer interaction, Natural (archaeology), Biology, Programming language, Paleontology