30079_14.pdf

CHAPTER 14

VIRTUAL REALITY—A NEW

TECHNOLOGY FOR THE

MECHANICAL ENGINEER

T\ishar H. Dani

Rajit Gadh

Department of Mechanical Engineering

University of Wisconsin—Madison

Madison, Wisconsin

14.1 INTRODUCTION

319

14.5.2 Desktop VR Systems

325

14.5.3 Hybrid Systems

325

14.2 VIRTUALREALITY

319

14.6 VRFORMECHANICAL

ENGINEERING

14.3 VRTECHNOLOGY

320

325

14.3.1 VR Hardware

320

14.6.1 Enhanced Visualization

325

14.3.2 VR Software

322

14.6.2 VR-CAD

325

14.4 VRSYSTEMARCHITECTURE 323

14.7 VIRTUAL PROTOTYPING/

MANUFACTURING AND VR

326

14.5

THREE-DIMENSIONAL

COMPUTER GRAPHICS vs. VR 324

14.5.1 Immersive VR System

324

14.1 INTRODUCTION

In recent times, the term virtual has seen increasing usage in the mechanical engineering discipline

as a qualifier to describe a broad range of technologies. Examples of usage include "virtual reality,"

"virtualprototyping," and "virtual manufacturing" In this chapter, the meaning of the term virtual

reality (VR) is explained and the associated hardware and software technology is described. Next,

the role of virtual reality as a tool for the mechanical engineer in the design and manufacturing

process is highlighted. Finally, the terms virtual prototyping and virtual manufacturing are discussed.

14.2 VIRTUALREALITY

The term virtual reality is an oxymoron, as it translates to "reality that does not exist." In practice,

however, it refers to a broad range of technologies that have become available in recent years to

allow generation of synthetic computer-generated (and hence virtual) environments within which a

person can interact with objects as if he or she were in the real world (reality). 1 In other instances,

it is used as a qualifier to describe some computer applications, such as a virtual reality system for

concept shape design or a virtual reality system for robot path planning.

Hence, the term by itself has no meaning unless it is used in the context of some technology or

application. Keeping in mind this association of VR with technology, the next section deals with

various elements of VR technology that have developed over the last few years. Note that even

though the concept of VR has existed since the late 1980s, only in the last two to three years has it

gained a lot of exposure in industry and the media. The main reason for this is that the VR technology

has become available at an affordable price so as to be considered a viable tool for interactive design

and analysis.

Mechanical Engineers' Handbook, 2nd ed., Edited by Myer Kutz.

Later, we will focus on VR applications, which allow such VR technology to be put to good use.

In particular, a VR-based application is compared to a typical three-dimensional (3D) computer-

aided-design (CAD) application to highlight the similarities and differences between them.

14.3 VRTECHNOLOGY

Typically, in the print media or television, images of VR include glove-type devices and/or so-called

head mounted displays (HMDs). Though the glove and HMD are not the only devices that can be

used in a virtual environment (VE), they do convey to the viewer the essential features associated

with a VE: a high degree of immersion, and interactivity.

Immersion refers to the ability of the synthetic environment to cause the user to feel as if he or

she is in a computer-generated virtual world. The immersive capabilities can be judged, for example,

by the quality of graphics presented (how real does the scene look?) or by the types of devices used

(HMD, for example). All VEs need not be immersive, as will become clearer from later sections.

Interactivity is determined by the extent to which the user can interact with the virtual world

being presented and the ways he or she can interact with the virtual world: for example, how the

user can interact with the VE (using the glove) and the speed with which the scene is updated in

response to user actions. This display update rate becomes an important ergonomic factor, especially

in immersive systems, where a lag between the user's actions and the scene displayed can cause

nausea.

With reference to the typical glove/HMD combination, the glove-type device is used to replace

the mouse /keyboard input and provides the interactivity, while the HMD is used to provide the

immersion. Though the glove and head-mounted display combination are the most visible elements

of a VR system, there are other components of a VR that must be considered. First, the glove and

HMD are not the only devices that can be used in a VE. There are many other devices in the market

that can be used for providing the 3D interactions capabilities. These are discussed in Section 14.3.1.

Second, the software in a VR system plays an equally important role in determining the behavior

of the system, is discussed. A wide variety of software tools for VR system are described in Section

14.3.2.

Third, the need for real-time performance, combined with the need to interface with a wide range

of devices, requires that special attention be paid to the architecture of a VR system. An example of

a typical VR system architecture is provided in Section 14.3.1.

14.3.1 VR Hardware

The hardware in a VE consists of three components: the main processor, input devices, and output

devices (Fig. 14.1). In the initial stages of VR technology development, in the 1990s, there was a

limited choice of computer systems that could be used for VR applications. Currently, all major

UNIX workstation vendors have specific platforms targeted to the VR market. These workstations

usually have a enhanced graphics performance and specific hardware to support VR-type activity.

However, with improvements in the processing speeds, of PCs, they are also becoming viable alter-

natives to more expensive UNIX-based systems. With prices much lower than their workstation

counterparts, these are popular with VR enthusiasts and researchers (with limited budgets) alike. The

popularity of the PC-based VR systems has spawned a whole range of affordable PC-based VR

interaction devices, some examples of which are provided in this section.

Main Processor

The main processor or virtual environment generator 2 creates the virtual environment and handles

the interactions with the user. It provides the computing power to run the various aspects of the

virtual world simulation.

The first task of the virtual environment generator is to display the virtual world. An important

factor to consider in the display process is the number of frames per second of the scene that can

be displayed. Since the goal of a VE is to look and feel like a real environment, the main processor

must be sufficiently powerful (computationally) to be able to render the scene at an acceptable frame

rate. A measure of the speed of such a processor is the number of shaded polygons it can render per

second. Typical speeds for UNIX-based Silicon Graphics machines range from 60,000 Tmesh/sec

(Triangular Mesh) for an Indigo2XL to 1.6 million Tmesh/sec for a Power Onyx/12. 3

The second task of the main processor is to interface with the different input and output devices

that are so important in providing the interactiveness in the VE. Depending on the platform used, a

wide range of input and output devices are available. A brief summary of such devices is provided

in the next two sections. Detailed description of such devices and hardware can be found in Ref. 4.

Input Devices

Input devices provide the means for the user to interact with the virtual world. The virtual world, in

turn, responds to the user's actions by sending feedback through various output devices, such as a

visual display. Since the principal objective of a VE is to provide realistic interaction with the virtual

Fig. 14.1 Hardware in a VR system.

world, input devices play an important role in a VR system. The mouse/keyboard interaction is still

used in some VR environments, but the new generation of 3D devices that provide the tools to reach

into the 3D virtual world.

Based on their usage, input devices can be grouped into five categories: tracking, pointing, hand-

input, voice-based, and devices based on bio-sensors. Of these, the first four types are typically used

in VR systems. Note that of the devices described below, only the devices in the first three categories

are used in VEs.

Tracking Devices. These devices are used in position and orientation tracking of a user's head

and/or hand. These data are then used to update the virtual world scene. The tracker is sometimes

also used to track the user's hand position (usually wearing a glove; see below) in space so that

interactions with objects in the 3D world are possible. Tracking sensors based on mechanical, ultra-

sonic, magnetic, and optical systems are available. One example of such a device is the Ascension

tracker. 5

Point Input Devices. These devices have been adapted from the mouse/trackball technology to

provide a more advanced form of data input. Included in this category is the 6-degree of freedom

(6-dof) mouse and force ball. The 6-dof mouse functions like a normal mouse on the desktop but as

a 6-dof device once lifted off the desktop. A force ball uses mechanical strains developed to measure

the forces and torques the user applies in each of the possible three directions. An example of force

ball-type technology is the SpaceBall. Another device that behaves like a 6-dof mouse is the Logitech

Flying Mouse, which looks like a mouse but uses ultrasonic waves for tracking position in 3D space.

Glove-Type Devices. These consist of a wired cloth glove that is worn over the hand like a

normal glove. Fiber-optical, electrical, or resistive sensors are used to measure the position of the

joints of the fingers. The glove is used as a gestural input device in the VE. This usually requires

the development of gesture-recognition software to interpret the gestures and translate them into

commands the VR software can understand. The glove is typically used along with a tracking device

that measures the position and orientation of the glove in 3D space. Note that some gloves do provide

some rudimentary form of tracking and hence do not require the use of a separate tracking device.

One example of such a glove is the PowerGlove 6 which is quite popular with VR home enthusiasts

since it is very affordable. Other costlier and more sophisticated versions, such as the CyberGlove,

are also available.

Biocontrollers. Biocontrollers process indirect activity, such as muscle movements and electrical

signals produced as a consequence of muscle movement. As an example, dermal electrodes placed

near the eye to detect muscle activity could be used to navigate through the virtual worlds by simple

eye movements. Such devices are still in the testing and development stage and are not quite as

popular as the devices mentioned earlier.

Audio Devices. Voice input provides a more convenient way for the user to interact with the

VE by freeing his or her hands for use with other input devices. Such an input mechanism is very

useful in a VR environment because it does not require any additional hardware, such as the glove

or biocontrollers, to be physically attached to the user. Voice-recognition technology has evolved to

the point where such software can be bought off the shelf. An example of such a software is

Voice Assist from SoundBlaster.

Output Devices

Output devices are used to provide the user with feedback about his or her actions in the VE. The

ways in which the user can perceive the virtual world are limited to the five primary senses of sight,

sound, touch, smell, and taste. Of these only the first three have been incorporated in commercial

output devices. Visual output remains the primary source of feedback to the user, though sound can

also be used to provide cues about object selection, collisions, etc.

Graphics. Two types of technologies are available for visual feedback. The first, HMD (head-

mounted display), is mentioned in Section 14.3. It typically uses two liquid crystal display (LCD)

screens to show independent views (one for each eye). The human brain puts these two images

together to create a 3D view of the virtual world. Though head-mounted displays provide immersion,

they currently suffer from poor resolution, poor image quality, and high cost. They are also quite

cumbersome and uncomfortable to use for extended periods of time.

The second and much cheaper method is to use a stereo image display monitor and LCD shutter

glasses. In this system, two images (as seen by each eye) of the virtual scene are show alternately

at a very high rate on the monitor. An infrared transmitter coordinates this display rate to the fre-

quency with which each of the glasses is blacked out. A 3D image is thus perceived by the user.

One such popular device is the StereoGraphics EyeGlasses system. 7

Audio. After sight, sound is the most important sensory channel for virtual experiences. It has

the advantage of being a channel of communication that can be processed in parallel with visual

information. The most apparent use is to provide auditory feedback to the user about his or her

actions in the virtual world. An example is to provide audio cues if a collision occurs or an object

is successfully selected. Three-dimensional sound, in which the different sounds would appear to

come from separate locations, can be used to provide a more realistic VR experience. Since most

workstations and PCs nowadays are equipped with sound cards, incorporating sound into the VE is

thus not a difficult task.

Contact. This type of feedback could either be touch or force. 8 Such tactile feedback devices

allow a user to feel forces and resistance of objects in the virtual environment. One method of

simulating different textures for tactile feedback is to use electrical signals on the fingertips. Another

approach has been to use inflatable air pockets in a glove to provide touch feedback. For force

feedback, some kind of mechanical device (arm) is used to provide resistance as the user tries to

manipulate objects in the virtual world. An example of such a device is the PHANToM haptic

interface, which allows a user to "feel" virtual objects. 9

14.3.2 VR Software

As should be clear from the preceding discussion, VR technology provides the tools for an enhanced

level of interaction in three dimensions with the computer. The need for real-time performance while

depicting complex virtual environments and the ability to interface to a wide variety of specialized

devices require VR software to have features that are clearly not needed in typical computer appli-

cations. Existing approaches to VR content creation have typically taken the following approaches 10 :

virtual world authoring tools and VR toolkits. A third category is the Virtual Reality Modeling

Language (VRML) and the associated "viewers" which are rapidly becoming a standard way for

users to share "virtual worlds" across the World Wide Web.

Virtual World Authoring and Playback Tools

One approach to designing VR applications is first to create the virtual world that the user will

experience (including ascribing behavior to objects in that world) and then to use this as an input to

a separate "playback" application. The "playback" is not strictly a playback in the sense that users

are still allowed to move about and interact in the virtual world. An example of this would be a

walk-through kind of application, where a static model of a house can be created (authored) and the

user can then visualize and interact with it using VR devices (the playback application).

Authoring tools usually allow creation of virtual worlds using the mouse and keyboard and without

requiring programs in C or C+ + . However, this ease of use comes at the cost of flexibility, in the

sense that the user may not have complete control over the virtual world being played back. Yet such

systems are popular when a high degree of user interaction, such as allowing the user to change the

virtual environment on the fly, is not important to the application being developed and when pro-

gramming in C or C++ is not desired. Examples of such tools are the SuperScape, 1 1 Virtus, 1 2 and

VREAM 1 3 systems.

VR Toolkits

VR Toolkits usually consist of programming libraries in C or C++ that provide a set of functions

that handle several aspects of the interaction within the virtual environment. They are usually used

to develop custom VR applications with a higher degree of user interaction than the walk-through

applications mentioned above. An example of this would be a VR-based driver training system, where

in addition to the visual rendering, vehicle kinematics and dynamics must also be simulated.

In general, VR toolkits provide functions that include the handling of input/output devices and

geometry creation facilities. The toolkits typically provide built-in device drivers for interfacing with

a wide range of commercial input and output devices, thus saving the need for the programmer to

be familiar with the characteristics of each device. They also provide rendering functions such as

shading and texturing. In addition, the toolkits may also provide functions to create new types of

objects or geometry interactively in the virtual environment. Examples of such toolkits include the

dVise library 1 4 the WorldToolKit library, 1 5 and Autodesk's Cyberspace Development Kit. 1 6

VRML

The Virtual Reality Modeling Language (VRML) is a relative newcomer in the field of VR software.

It was originally conceptualized as a language for Internet-based VR applications but is gaining

popularity as a possible tool for distributed design over the Internet and World Wide Web.

VRML is the language used to describe a virtual scene. The description thus created is then fed

into a VRML viewer (or VRML browser) to view and interact with the scene. In some respects,

VRML can be thought of as fitting into the category of virtual world authoring tools and playback

discussed above. Though the attempt to integrate CAD into VRML is still in the initial phase, it

certainly offers new and interesting possibilities. For example, different components of a product

may be designed in physically different locations. All of these could be linked together (using the

Internet) and viewed through a VRML viewer (with all the advantages of a 3D interactive environ-

ment), and any changes could be directed to the person in charge of designing that particular com-

ponent. Further details on VRML can be found at the VRML site. 1 7

14.4 VR SYSTEM ARCHITECTURE

To understand the architectural requirements of a VR system, it will be instructive to compare it with

a standard 3D CAD application. A typical CAD software program consists of three basic components:

the user input processing component, the application component, and the output component. The

input processing component captures and processes the user input (typically from the mouse/key-

board) and provides these data to the application component. The application component allows the

user to model and edit the geometry being designed until a satisfactory result is obtained. The output

component provides a graphical representation of the model the user is creating (typically on a

computer screen).

For a VR system, components similar to those in CAD software can be identified. One major

difference between a traditional CAD system and a VR-based application system is obviously the

input and output devices provided. Keeping in mind the need for realism, it is imperative to maintain

a reasonable performance for the VR application. Here "performance" refers to the response of the

virtual environment to the user's actions. For example, if there is too much lag between the time a

person moves his or her hand and the time the image of the hand is updated on the display, the user

will get disoriented very quickly.

One way to overcome this difficulty is to maintain a high frame rate (i.e., number of screen

updates per second) for providing the graphical output. This can be achieved by distributing the input

processing, geometric modeling, and output processing tasks amongst different processors. The reason

for distributing the tasks is to reduce the computational load on the main processor (Fig. 14.2).

Typical approaches adopted are to run the input and output processing component on another

processor (Windows-based PC or a Macintosh) while doing the display on the main processor. In

addition to reducing the computational workload on the main processor, another benefit of running

the input component on a PC is that there are a wide variety of devices available for the PC platform,

as opposed to the UNIX platform. This also has an important practical advantage in that a much

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika: