Route 20, Autobahn 7, and Slime Mold: Approximating the Longest Roads in USA and Germany With Slime Mold on 3-D Terrains.

A cellular slime mould Physarum polycephalum is a monstrously large single cell visible by an unaided eye. The slime mold explores space in parallel, is guided by gradients of chemoattractants, and propagates toward sources of nutrients along nearly shortest paths. The slime mold is a living prototype of amorphous biological computers and robotic devices capable of solving a range of tasks of graph optimization and computational geometry. When presented with a distribution of nutrients, the slime mold spans the sources of nutrients with a network of protoplasmic tubes. This protoplasmic network matches a network of major transport routes of a country when configuration of major urban areas is represented by nutrients. A transport route connecting two cities should ideally be a shortest path, and this is usually the case in computer simulations and laboratory experiments with flat substrates. What searching strategies does the slime mold adopt when exploring 3-D terrains? How are optimal and transport routes approximated by protoplasmic tubes? Do the routes built by the slime mold on 3-D terrain match real-world transport routes? To answer these questions, we conducted pioneer laboratory experiments with Nylon terrains of USA and Germany. We used the slime mold to approximate route 20, the longest road in USA, and autobahn 7, the longest national motorway in Europe. We found that slime mold builds longer transport routes on 3-D terrains, compared to flat substrates yet sufficiently approximates man-made transport routes studied. We demonstrate that nutrients placed in destination sites affect performance of slime mold, and show how the mold navigates around elevations. In cellular automaton models of the slime mold, we have shown variability of the protoplasmic routes might depends on physiological states of the slime mold. Results presented will contribute toward development of novel algorithms for sensorial fusion, information processing, and decision making, and will provide inspirations in design of bioinspired amorphous robotic devices.