作者: Philip Klahr
DOI: 10.1016/B978-0-408-01504-2.50014-4
关键词:
摘要: We present a new methodology for the design and development of large-scale simulations. Such simulations have often been criticized their lack intelligibility, modifiability, credibility, performance. They are costly to build, run, interpret, change. Recent advances in artificial intelligence, most notably areas expert systems object-oriented languages, suggest approaches overcoming these limitations. applied this technology implementation simulation language environment called ROSS (Rule-Oriented Simulation System). Within our research system, we implemented two military battle simulations, SWIRL (a strategic simulation) TWIRL tactical simulation), which simulate interactions outcomes combat between opposing forces. A graphics facility has developed visually display runs dynamically. written Franzlisp running on VAX. Our routines C loaded directly into environment. processor is an AED 512. Also experimenting with network Xerox 1100s (Lisp Machines) work distributed (discussed elsewhere [4]).