Tracking Systems

Interaction with an immersive environment should feel natural and transparent to the user. The two main issues are the tracking of the head position and orientation and certain other body parts, such as the hand and fingers. There are six possible technologies for tracking systems.

Electromagnetic (EM) trackers have a transmitter that emits Electromagnetic (EM) field along three orthogonal axes that are detected by sensors. The sensors report information about their position and orientation with respect to the source. There are two problems with EM systems. The first is latency, which is the time lag between a sensor movement and the time it is reported. Current systems have a latency of about 0.1 sec. Another problem is accuracy, EM trackers are very sensitive to the presence of metal and become unreliable. An advantage to these trackers is that they can be freely moved and are not perturbed by non-metallic objects, such as the user's body. An example is the Polhemus Isotrack.

Mechanical trackers, such at the FSL Boom, have a rigid structure with several joints. One end is attached to the object to be tracked and the other end is fixed in place. The joint angles are measured and reported to give the objects position and orientation. There is a very small latency and high accuracy for these systems. A disadvantage is restriction of movement.

Acoustic trackers use ultrasonic sound. A source produces pulses that are received by a set of microphones usually arranged as a triangle. The time each pulse reaches the different microphones gives the source position and orientation. They have the same latency problem as EM trackers, and while not affected by metal, they are affected by anything that comes between the source and the microphones, such as a body part.

Optical trackers use a combination of LED's, video cameras, and image processing. The LEDs can be placed on the object to be tracked while the cameras are fixed or the cameras are placed on the object and there is an array of LEDs in the ceiling. The position and orientation are obtained from signal processing techniques. These are affected by intervening objects. An example of the second type is an experimental system at the U. of North Carolina.

The Image Processing technique uses video cameras to capture images of the users. Image processing methods can be used to identify different body parts such as head, arms, or leg. Currently these are only used in 2D applications by Myron Krueger.

Inertial systems use gyroscopes to measure the three orientation angles. They are connected by a cable to the computer system so are limited only by this. These are not yet commercially available and are still experimental.


Main Virtual Environments Page
HyperVis Table of Contents

Last modified on February 18, 1999, G. Scott Owen, owen@siggraph.org