Holly Adams

Abstract

Virtual Reality is a completely immersive computer generated environment which allows for user interaction. A computer and associated tracking peripherals are used to generate a realistic scene and update it based on the position and orientation of the user. VR systems are individual in their ability to provide the user with a 360 degree field of view in all orientations. Since it originated over 30 years ago, virtual reality(VR) has suffered from a problem known as lag. Lag is the system's inability to keep up with the user's actions within a virtual environment. Lag occurs for many reasons. Anything from slow processing speeds of the tracker or the computer to slow data transfer between the computer and tracking peripherals will cause an increase in lag. Even if the tracking peripherals could provide information to the computer generating the virtual environment immediately, lag would still be an issue. It is not until after the computer generating the virtual environment receives information from the tracking peripherals, that the computer begins to generate the environment. On average, it takes approximately 16ms to generate a virtual environment. Under ideal conditions, this would create a 16ms delay. Actual environment generation time is dependent on the speed of the central processing unit or computer generating the environment and varies from system to system. For this reason, it would be advantageous to have a tracking system which could predict the user's actions beforehand. Prediction would allow the system to begin generating a new scene within the environment and display that scene at the appropriate time rather than several milliseconds after the fact. Both inertial and magnetic tracking systems are currently used in VR settings, but neither provides the speed and quality necessary to maintain a realistic experience within the environment. InterSense, a new company in the Boston area, recently released a hybrid tracking system which they claim surpasses the standard magnetic tracker on the market. This system, the IS600, combines inertial and acoustical information to maintain 6 degrees of freedom. The IS600 reports yaw, pitch, and roll, as well as x, y, and z position information. In order to determine the success of this system, it was necessary to characterize the system performance and then integrate it into a virtual environment for perceptual testing. Characterization of the IS600 revealed failures of the system at high and low angular velocities and a random sampling rate. The system's inertial prediction was successful and very effective for smooth motions. Thirteen subjects were tested to determine their preference of prediction within a virtual environment. The subjects were asked to choose between environments generated with 30ms of inertial prediction and environments generated without prediction. Results were not sufficient to conclude that prediction was effective, but this test can not be used as an accurate measure of the system's performance. Other problems, such as the random sampling rate, of the system may be the cause of these inconclusive results. Additional testing will be necessary to determine the effectiveness of the product.

Publication Date

1998

Document Type

Thesis

Advisor

Not listed.

Comments

Note: imported from RIT’s Digital Media Library running on DSpace to RIT Scholar Works in February 2014. Senior project. Physical copy available through RIT's The Wallace Library at: TR395.W33

Campus

RIT – Main Campus

Download

Share

COinS