This paper focuses on the problem of orchestrating the movements of bot agents in the videogame Quake 3 Arena. Since the specific patterns of movement that one may wish to enforce may be various, and serve different purposes (have bots meet somewhere, move in formation, or surrounding human players), a general and flexible approach is required. In this paper we discuss how the Co-Fields coordination model can be effectively exploited to this purpose. The key idea in Co-Fields is to model the agents' environment by means of application-specific computational force fields, leading agents' activities to a globally coordinated and adaptive motion behavior. The Co-Fields model is described both in general terms and in the specific Quake 3 Arena implementation, and several application examples are presented to clarify it. Also, the paper outlines the general applicability of the approach besides the Quake scenario and in areas such as mobile computing and mobile robots.
Motion coordination in the Quake 3 Arena environment: A field-based approach / Mamei, Marco; Zambonelli, Franco. - STAMPA. - 3374:(2005), pp. 264-278. (Intervento presentato al convegno First International Workshop on Environments for Multi-Agent Systems tenutosi a New York, NY, usa nel JUL 19, 2004) [10.1007/978-3-540-32259-7_14].