SGER: Preliminary Studies of Cellular Automata Self-Replication Programming

Iowa State University, Ames IA

Investigators

Abstract

The PI will investigate cellular automata self-replication programming, to pave the way for more complex studies of this subject in the future. In preliminary work, the PI has been able to construct systems where self-replicating structures emerge in a randomly initialized space, and also systems where self-replicating structures can solve a restricted version of the satisfiability (SAT) problem. Both models take advantage of the massively parallel nature of the underlying cellular automata space within which structures replicate and evolve. This has led the PI to conjecture that self-replication programming may constitute a new paradigm for developing massively parallel algorithms to solve computationally expensive problems. With this long term goal in mind, the PI will study during the coming year two relatively easier subgoals: the design of compact self-replicating structures where more data for computation can be stored in the same size cellular space than currently possible, and new bisexual self-replicating models that incorporate the crossover operator from genetic algorithms. To provide the requisite testbed for this work, the PI will utilize the Java Virtual Machine to extend his high level Trend cellular automata programming language and simulation environment to a variety of computer platforms.