Morphogenesis as a Model
We have an example of how hierarchical self-assembly can be accomplished in embryological morphogenesis, which coordinates billions or trillions of cells to assemble a complex, hierarchical body (Nüsslein-Volhard, 2008). Even a relatively simple animal has a large number of tissues, organs, vessels, nerves, etc. that are physically structured in a complex and functional organization. Moreover, multicellular organisms are hierarchically organized from the cellular (and indeed nano) level up to macroscopic level. Beginning from a single cell, the developing zygote begins to organize itself, establishing poles and layers, and the progressing organization governs future development, so that the microscopic agents (the cells) create the structure that governs their own future behavior. Cells migrate, following chemical gradients, and create forces and pressures that help to shape the tissues. Under the influence of structured signals, cells differentiate into functionally distinct tissues. Thus, the development of the embryo provides an inspiring example of how microscopic agents can coordinate their mutual behavior so as to self-organize into an immensely complicated structure. Our goal in artificial morphogenesis is to mimic these processes for the self-assembly of complex, hierarchical artificial systems by massive robot swarms.