The Newtonian Architecture for Virtual Landscapes : an architecture, model and implementation

Date

2017-12-08

Authors

Wedlake, Martine Bruce

Journal Title

Journal ISSN

Volume Title

Publisher

Abstract

There is much research in the literature regarding the construction of distributed virtual reality implementations. After evaluating some well-known virtual reality systems, it was determined that several problems exist that need to be solved. In particular: network efficiency, object distribution and coherency, inadequate system resource management, and overall performance. In order to properly address these issues, a holistic design approach is taken. The entire system is examined, rather than focusing on a specific problem area (such as the human-computer interface). The major component of this work, the Newtonian Architecture for Virtual Landscapes (NAVL), is presented to respond to the problems areas discovered. Highlights of the architecture include: (1) A distributed client/server network that addressed the networking issues. (2) Autonomous objects encapsulate control and object state into a single entity. Using autonomous objects avoids lengthy synchronization processes (e.g., full database locking). (3) ForceLets, a novel synchronization method, minimize the network bandwidth required to keep an object synchronized at remote locations. In addition, ForceLets provide much improved synchronization of the object at the remote locations in the presence of network lag. Implementation details of the NAVL prototype are also presented. The implementation consists of an object simulation and execution unit, rendering and collision detection unit, and network subsystem and protocols. An evaluation of the NAVL system architecture examines the efficiency of the key architectural components: (1) A bandwidth and latency analysis examines the efficiency of the distributed client/server network. (2) The object distribution and coherency components are tested directly from the prototype. Profiles of actual prototype execution are used to show the efficiency gains of the ForceLet approach as compared to the commonly used stream-of-data coherency mechanism. (3) The rendering and collision detection unit is tested by examining the effects on CPU utilization and frame rate with increases in the number of virtual objects.

Description

Keywords

Human-computer interaction, Virtual reality

Citation