Euclidean TSP using characteristics of slime mold

This paper reports one engineering application of spatial searching characteristics of slime mold (multi-cellular) amoeba for autonomous parallel distributed processing systems. The remarkable characteristic of slime mold is that it transforms from unicellular period to multicellular period according to the state of the environment. Such characteristics present one idea for the difficulty of autonomous PDP system, such as environmental recognition, cooperative behaviors, and self-organized system control. This paper proposes the Amoeba Searching Behavior Model, and applies it to the Euclidean TSP as an example of spatial searching environment. The proposed model consists of a new field named as the Vibrating Potential Field and a new parameter tuning method inspired from thermo-dynamics. The computational application for Euclidean TSP shows advantages of the model and derives powerful characteristics for intelligent searching mechanism: these are heteronomous behavior and autonomous behavior