COntext-aware multi-agent systems negotiating intensive fields

Abstract

This paper presents research and experimentation with context-aware multi-agent-based design systems to simulate and propose urban schemes that specifically utilize fields of differentiated intensity data in order to propose an infrastructure to support urban revitalization. The research objective of the work is to propose a semi-autonomous system that can be deployed to analyze and then design and construct urban-scale infrastructure where regeneration or revitalization is necessary, emphasizing the possibilities for emergent efficiencies. The research presents the system development in the context of a hypothetical design research for a contaminated area of Mexico City. The problem definition, system development and simulation and analysis of the results to date are described and discussed. The system consists of a series of steps undertaken by a group of hypothetical simulated robotic agents that react to local conditions to generate a network for irrigation and buildable areas in a contaminated brown-field site with high salinity. Evaluation of the system is conducted by comparing initial data sets with the final form of the simulated irrigation network. These results provide a workflow for negotiating with intensive fields generally. The work also illustrates a site-specific scheme for deployment of different infrastructure networks. The work is discussed in the context of generative urban design and the ways in which architects are appropriating simulation technologies to generate urban form.