Parallel Accelerated Virtual Physarum Lab Based on Cellular Automata Agents

摘要: Self-aware and self-expressive physical systems are inspiring new methodologies for engineering solutions of complex computing problems. Among many other examples, the slime mold Physarum Polycephalum exhibits self-awareness self-expressiveness while adapting to changes in its dynamical environment solving resource-consuming problems like shortest path, proximity graphs or optimization transport networks. As such, modeling mold’s behavior is essential when designing bio-inspired algorithms hardware prototypes. The goal this paper combine one powerful parallel computational tools, cellular automata (CA) with adaptive potential mold. Namely, we propose a CA model multi-agent approach imitate plasmodium. We then test efficacy proposed on graph such as maze problem traveling salesman (TSP). Finally, virtual evaluated data set pattern recognition purposes achieves form very effectively letters alphabet, especially compared real experiments performed prove model. Furthermore, exploit CA’s inherent parallelism make model’s responses faster, both GPU implementations compared. result, an accelerated lab developed which uses describe plasmodium can be used intelligent, autonomous, self-adaptive system various heterogeneous unknown environments spanning from different types up life-time applications.