Artificial Neural Path Formation by Massive Microrobot Swarm

Artificial morphogenesis uses processes inspired by embryology to control massive swarms of robots to assemble complex physical structures. This video shows several simulated applications of a modified flocking algorithm for collision-free routing of neural fiber bundles from origins to destinations. Flocks follow an increasing gradient of an attractant morphogen diffusing from the destination. Existing paths either emit a repelling morphogen or absorb the attractant, thus steering flocks around them. Treating flocks as continuous masses facilitates scaling to arbitrarily large flocks of microrobots.

Representative citations:

MacLennan, B. J. (2018). “Coordinating Swarms of Microscopic Agents to Assemble Complex Structures,” Ying Tan (Ed.), Swarm Intelligence, Vol. 1: Principles, Current Algorithms and Methods (PBCE119), ch. 20, pp. 583–612, IET (Inst. Eng. & Tech.).

MacLennan, B. J. (2019). “A morphogenetic program for path formation by continuous flocking.” International Journal of Unconventional Computation 14, 91–119.