The aggregation phenomenon resulting from general taxis stimulus is studied here using an agent-based approach. Usually described by a set of partial differential equations in a mathematical formalism, this phenomenon is studied in this paper with a system of "agents-particle" derived from particle methods. This approach allows us to simulate the phenomenon in an individual and discrete way and then to enrich the mathematical description in terms of interaction specifications and model explorations. This paper thus establishes the basis of an agent-oriented model able to handle more specific multi-level issues in general taxis processes and in the resulting aggregation phenomenon. Formally built around two basic dynamics - spatial diffusion due to density gradient and attraction following a gathering signal - our model shows, in a first step, the emergence of lifelike aggregates from a randomly distributed population. In a second step, it proceeds, using a Hopfield network, to the detection and reification of those emerging structures in order to be able to deal with the resulting level of complexity. © 2012 Published by Elsevier Ltd.
Dembele, J. M., & Cambier, C. (2012). An agent-particle model for taxis-based aggregation; emergence and detection of structures. In Procedia Computer Science (Vol. 9, pp. 1484–1493). Elsevier B.V. https://doi.org/10.1016/j.procs.2012.04.163