Design and implementation of distributed virtual geographic environment system based on MAS

Abstract

ABSTRACT The use of distributed virtual geographic environments (DVGE) is one of the most promising uses of virtual reality in geographic field. Multi-Agent Systems (MAS) is flexible and adaptive enough to resolve highly complex distributed computing problems. This paper aims to discuss how to apply MAS technology to improve the efficiency of collaborative work among multi-user. We firstly design the th ree-layer system architecture and multi agent system of the DVGE. Then dynamic cooperative group algorithm is also addressed. Finally, we built a collaborative virtual environment and implemented prototype system in silt dams system planning on a case study area, Jiu-Yuan-Gou watershed of Loess Plateau, China. Experiment results prove that the scheme addressed in this paper is efficient and feasible. Keywords: Distributed Virtual Geographic Environment, Multi Agent System (MAS), Dynamic Cooperative Group Algorithm 1. INTRODUCTION As an interdisciplinary of visualization and GIS, virtual geographic environment (VGE) was proposed in 1999. Unlike the traditional data-centered GIS, VGE is a human-centered environment that represents and simulates geographic environments (both physical and human environments). It allows geographically distributed multi-user to visually explore spatial information, design models, implement complex computation, and to provide support for decision-making [1]. A distributed virtual geographic environment (DVGE) system is a virtual Internet-based 2D and 3D environment that provides users with shared space and a collaborative platform for publishing multidimensional geo-data, and for simulating and analyzing complex geo-phenomena. Users logging into the system can be from different clients, which are often not in the same geographic area, but can nevertheless share distributed geo-information resources, including geo-data and geo-models, and can also complete collaborative tasks [2]. As a new research direction, the study of the DVGE involves diverse aspects including participants and organizations, mediated tools, geo-problem contexts, and supportive environments and so on. How to share these distributed and heterogeneous geographic data and offer more rapid and convenient services is a very important task. Designing and developing a DVGE system based on new computer technology is a highly relevant pr oblem for computer experts and geo-science experts [1, 2]. An agent is an autonomous computer software object that has specific functions and responds to specific events, has the capability to make decisions and take actions to satisfy internal goals based on its perceived environments and predefined knowledge rules or user designated instructions and can interact with other agents and a common environment. “A Multi-agent Systems can therefore be define d as a collection of possibly heterogeneous, computational entities, having their own problem solvin g capabilities and which are able to interact among them in order to reach an overall goal”[3]. Multi-Agent Systems (MAS) have been traditi onally employed in Artificial Intelligence to solve highly complex distributed planning problems. Today, this technology is increasingly applied to different information systems. The MAS technology is flexible and adaptive enough to solve different distributed problems in the DVGE such as scheduling, web search, and resources allocation [4]. In fact, the multiple fields and cross-disciplines research are very necessary to resolve the various geographic problems in the DVGE. In relevant work, Manoharan established a collaborative urban planning prototype system to assist shared analysis of urban planning proposals by visualization and interaction with spatial data [5]. MacE achren addressed geospatial virtual environments (GeoVE) and dialogue-assisted