VI. CONCLUSION
In this paper, we presented a hardware-based maze-solving approach via an electronic model of the oscillatory internal motion mechanism of plasmodium. The efficiency and generality of the proposed electronic circuit was validated through SPICE-level simulations compared with data from two different biological experiments, namely enhancing of Physarum’s protoplasmic tubes along shortest path [3], and chemo-tactic growth of slime mould [5]. The proposed circuit is operated in the region of maximum adaptation (learning), so every module adjusts its function depending on the characteristics of the input waveform, whereas network expansion corresponds to gradual training of subsequent interconnected modules. The introduction of inherent noise abilities to the equivalent biological circuit resulted in several (including the optimal) maze solutions which is closer to non-deterministic real organisms’ behavior [3], [5]. Future work could include modifications to the proposed circuit modules so as to support also the shrinking stage of the vascular network of the plasmodium after the shortest possible interconnection network has been computed. The series connection of distributed identical modules with circuit parameters adjusted for specific transmission characteristics resembles the distributed element model of transmission lines. Therefore, the proposed here circuit could be optimized to minimize circuit overhead by stages b–e through optimal parameter value selection for perfect signal transmission and reflection to achieve self-reinforcement of the solution paths. Other extensions of the proposed work could include the circuit modeling of the behavioral characteristics of plasmodium, demonstrated in a variety of biological experiments concerning the solution of several other problems such as the traveling salesman or the execution of logic computations.
