IIMS 94 contents
[ IIMS 94 contents ]

Virtual reality comes of age

Robert J. Stone
Advanced Robotics Research, Salford, England
As one of Europe's leading organisations in Virtual Reality (VR) endeavours for industrial applications, ARRL has faced two key challenges during its transition from a UK Government supported institution to a commercial contract research company, particularly at a time of national and international recession. These were (a) the fact that very few industries have been in a position to sub-contract VR work out (eg, for consultancy, feasibility study, or turnkey system agreements) and (b), of those that ARRL recognised as having serious applications, most still considered VR technology to be restricted to games and entertainment arcades, and of no possible use to their business. During the closing months of 1992 and early in 1993, a number of events came about which changed this situation dramatically. This paper briefly reviews the underlying technologies associated with VR and goes on to describe how ARRL has "sold the vision" of VR to a sceptical and money conscious industry. The paper also describes the early stages of the result of a nationwide "sales" exercise - the first Virtual Reality and simulation (VRS) project in Europe to date, funded solely by British companies and academia.


Introduction

On a February morning in Monte Carlo in 1993, a member of the French clergy apprehensively donned a helmet mounted display and grasped a free flying joystick. At the same time a young lady in Paris indulged in a similar activity. Looking to his right, the priest was greeted by a computer generated cartoon like character who, speaking with a soft French accent, proceeded to provide a guided tour of a graphical reconstruction of the Cluny Abbey, a building that has not existed for a number of centuries. Such was the first European demonstration of a shared virtual experience, with two people communicating visually and vocally over a French Numeris ISDN telecommunications link.

Very few "new" technologies have captured the imagination of people in all walks of life as much as Virtual Reality (VR) - the ability to immerse human beings in an alternative, three dimensional computer generated environment. It seems that not a week goes by without some mention in the newspapers and on the television of new breakthroughs in VR for entertainment, medicine and defence. VR has been used to sell male perfumes, clothing and computer games, to cure the fear of flying by creating the illusion that passengers were actually on a train[1]... even the UK's BBC 9 O'clock News now sports a so called "virtual" studio.

By the mid 1980s, only a small handful of researchers outside of the USA were aware of the significant developments in telepresence for robotic space applications and extravehicular activity training that were occurring in NASA's human factors laboratories. Of those who were, even fewer could predict that the subsequent commercialisation of a pair of helmet mounted, low resolution, liquid crystal displays and a Lycra glove, instrumented with fibre optics, would cause such a technological revolution throughout the United States and, 3-4 years later, in Europe. Virtual Reality (Anon., 1991; Rheingold, 1991; Krueger, 1991) suddenly provided the international robotics research community with a reasonably low cost technology base from which telepresence could, at last, be investigated and implemented. Who else could have further predicted that the same technology would spawn products based essentially around a couple of electromagnets, a pair of pocket televisions and a cannibalised PC joystick, thereby enabling two British companies to expedite a multi-million pound flotation on the UK's Stock Exchange in 1993?

Something old, something new

In contrast to the understanding of most of the general public and academia (certainly in Europe), VR is not really as new as one might first think. In fact, the birth of VR, as a means of providing a user with an alternative reality or personal experience can be traced back to the early 1950s, courtesy of a system invented by an American - Morton Heilig - called "Sensorama". Stimulated by the wide screen Cinerama films being shown by Hollywood, Heilig's Sensorama brought the sensation of being "part of the action" to a single viewer. For just a dime, members of the public could experience Brooklyn drive throughs or Turkish belly dancers, and, in addition to the canned film sequences, have their senses stimulated by perfumed air jets and a vibrating seat.

The concept of immersive VR (although originally not referred to as such) originated in the early 1960s. Its emergence is in part attributed to the work of Ivan Sutherland. His first system, called Sketchpad, was a computerised design tool which allowed users to design and manipulate graphical objects on a screen using a light pen. Sketchpad ran on a TX-2, the most powerful computer available at that time. It was not only a tool for visualisation, it was an operating environment in its own right. Sutherland's later work, described in The Ultimate Display (Sutherland, 1965), was based on research for the US Advanced Research Projects Agency, ARPA (which has recently, yet again, dropped its first letter of D - for Defence). Using a ceiling mounted, cantilevered stereoscopic display headset boasting a 400 field of view - the Sword of Damocles - primitive wire frame graphics with hidden line removal could be displayed to the wearer in stereo, or 3D.

Most of the key developments in the evolution of VR occurred in the 1980s, through such defence initiatives as the Supercockpit Project, based at the Wright Patterson Air Force Base in Dayton, Ohio, through efforts in telepresence and robot control, at the NASA Ames Research Centre in California (eg, Fisher et al, 1988), and through the enthusiastic world wide marketing campaign of the US Company, VPL Inc (for a more detailed and reasonably accurate history, see Pimental & Teixeira, 1993). VR first came to the notice of the British and European public late in 1990, when individuals who had been involved in developing the technology on the Eastern side of the Atlantic, presented their work at the London Computer Graphics Conference. One of the prime movers at that time was, without doubt, W Industries based in Leicester. The emergence of their Virtuality [2] Games Systems (and the subsequent spread of "Location Based Entertainment Centres" across the UK) was accompanied not only by Hollywood style product launch events, but also by much speculation and hype about the role VR would play in the future of computing technology.

If it ain't immersive, it ain't VR

Despite the propaganda dispersed by the "headset and glove" brigade, VR is concerned first and foremost with intuitive interaction with real time, three dimensional, computer generated worlds. Two complementary forms of VR exist, referred to as "desktop" and "immersive". Desktop VR allows a user to build and interact with virtual worlds using a range of computing equipment, from personal computers to powerful graphics workstations. One important desktop VR system, based around a 486 PC with graphics accelerator, has been developed by the UK Company Dimension International, of Aldermaston. Dimension's Virtual Reality Toolkit (VRT), known as Superscape, allows a VR developer to construct all the elements of a virtual environment, using software akin to that used for computer aided design, and locate them intelligently in the environment "shell". The VRT software supports the user in organising the virtual elements in such a way that the power of the PC can be used to its full potential - down loading more detailed versions of an object as the user moves closer, "switching" elements (eg, whole rooms) on and off as the user moves from one part of the virtual environment to another. The user views the virtual world via a high resolution monitor, moving and manipulating objects using a standard mouse and Spaceball 6 degree of freedom (X, Y, Z; roll, pitch, yaw) strain gauge hand controller. VRT was in fact used to provide the computer generated imagery for the so called "first" VR TV Game - the UK BBC's Cyberzone. Other packages exist, primarily American in origin, with a full system price range of between 2000 and 14,000.

Strictly speaking, immersion is more of a "purist's" concept of VR, the term being typically used to describe the experience of "being there" by users sporting head mounted displays and interactive gloves. The main goal of immersive VR is to exploit the natural human skills of the user - head and limb movement, gestures, stereoscopic vision, hearing and speech, in order to get him or her "closer" to the computerised application. In other words, users should not be restricted in their use of computer technology by the need to be fluent in a particular programming language or computing skill. They should, courtesy of immersive VR, be able to interact with the system in the same way they interact with objects and events on a day to day basis. Whilst for many applications, this is an applaudable aim, it is one that VR has not, to date, been able to achieve successfully.

The critical area of concern centres on the design and image quality of current generation head mounted displays. Until very recently, immersive VR users had to contend with cumbersome helmets housing a pair of low resolution (eg, 442 (horizontal) x 238 (vertical) pixels) liquid crystal displays (LCDs) - in effect, cannibalised pocket televisions - and weighty optics. Not only have headsets been the one component of VR which has hampered the wider uptake of the technology in applications other than entertainment, they have also been the source of much concern from a health and safety point of view. The quality of displayed images in these systems has, to date, been so poor that US research has classified the visual world presented to immersed VR users as that experienced by individuals who are, in reality, legally blind! Other symptoms such as visual fatigue, disorientation and nausea have been reported by many users (both during and after immersive VR sessions), yet, in stark contrast to the legislation that exists for work involving conventional visual display units, nothing has yet been done to control or restrict the exposure of the public to this primitive technology. With recent concerns for the health and safety of video games users, this situation is likely to be the focus of much media attention when, later this year (and assuming there are no further setbacks), Sega launches its immersive VR add on to the Genesis System in the USA. The British Mega CD equivalent is likely to be available by the Summer of 1994, at a cost of less than 200.

Despite these problems, the stereoscopic headset situation is improving gradually. Efforts in the USA and Europe are now concentrating on the use of miniature cathode ray tubes (CRTs) which, together with new LCD shutter devices to generate colour, offer resolutions of the order of 1280 x 960. Nearly 20 LCD/CRT headset systems will come into existence over the next year or so. Whilst the search for new and better quality headset designs is to be applauded, the transition from LCD to the higher energy CRT devices, mounted around the head (eg, close to the temporal lobes of the brain), together with their typical arrangement of narrow field of view optics, have inevitably opened up a whole new set of health and safety problems.

Of course, headsets are not the only method by which to view a virtual world. Other, more specialised display systems include multiple TV monitors (as used in some "Location Based Entertainment" VR Centres and on the TV program Cyberzone, mentioned earlier), counterbalanced monoscopic and stereoscopic viewers (eg, Fake Space Labs/Division Inc's BOOM - Binocular Omni-Orientable Monitor), liquid crystal shuttered ("field sequential") spectacles, large screen 2D/3D video projectors and passive stereoscopic techniques (ie, using polarised filters viewing glasses).

As far as the control of such activities as motion and manipulation in a virtual world are concerned, there are numerous commercial devices now available, ranging in cost from a few tens to many thousands of pounds. The most common form of device used in immersive VR consists of an electromagnetic spatial tracker embedded within a low cost hand grip or modified PC games joystick (a similar tracker is fitted to VR helmets to register head movement). These hand grips have become popular over the last 12-18 months, due to the high cost and low reliability of glove based products. As with headsets, fortunately, new, cheaper and more robust glove products are beginning to emerge. One such product - a glove using conductive elastomer sensors - forms part of a catalogue of instrumented clothing for robotics, animation and VR applications and has been developed by the Cardiff based Company TCAS Effects Ltd. Tactile and force feedback systems, together with replication of other skin mediated senses, still present the VR Community with a major challenge, although there are considerable efforts across the globe addressing proportional pneumatics, exoskeletons, vibratory transducers, shape memory alloys and electromechanical back driveable joysticks.

One potentially exciting area in the field of data input systems design for VR is that of "Biocontrol". The term biocontrol refers to the recording of bioelectric or electrophysiological signals, originating naturally from processes within the human body (eg, muscle and nerve activity), which, suitably filtered and amplified, can be used to control other processes, remote from the originating human body (internal bodily processes can also be controlled to some extent, as with biofeedback training). Until quite recently, biocontrol has been believed to be of interest to followers of science fiction. However, due to relatively new developments, originally designed for seriously handicapped individuals, biocontrol is now being investigated in laboratories researching virtual reality and telepresence.

Returning briefly to the subject of computing platforms for coordinating the display of stereoscopic images, and for integrating and coordinating the data generated by VR control peripherals, over the past 12 months or so, the VR Community has received a significant boost by the emergence of a new breed of graphics "supercomputers", capable of permitting real time interaction with near photorealistic virtual worlds. It has been known for some time that Silicon Graphics (SG) would eventually target the Virtual Reality "market" with a professional product that would compete seriously in terms of performance and price with existing platforms and produce virtual images of acceptable quality to industry, especially when combined with the model building and run time virtual world generation power of software packages, such as ModelGen/MultiGen and Sense8's WorldToolKit. SG's Crimson RealityEngine and, more recently, the RealityEngine2, or Onyx supercomputer did just this, with many of the functions provided in software form on other platforms (eg, texture mapping, anti-aliasing) being integrated on dedicated hardware within the SG system. This provided a capability to be able to handle much larger and more visually complex worlds - with a stereoscopic display and potential multi-user capability.

Other companies have attempted to produce VR computing systems or dedicated add on boards, but have met with limited success, some even collapsing, thereby leaving many purchasers of graphics boards with non-standard, non-upgradeable hardware and software. Many companies seem to be obsessed with pushing their hardware and/or software product to become the "European Standard". If there is one thing the VR Community does not need yet, it is standards for operating systems, toolkits and hardware. Data interchange protocols, yes (eg, the European ISO/ESPRIT STEP Initiative); human factors standards, definitely; but not systems standards forced upon the R&D community through commercial greed rather than technical rationale. Until the volatile VR Market actually settles down (and many believe this Will not happen in Europe until post 1995), there can be no guarantee that any one product line - hardware or software - will emerge as the de facto standard. The international VR product range - and therefore the "state of the art" - is changing and growing all the time. For many, if not all, VR applications outside the leisure and entertainment industry, it is not satisfactory to consider just one company's system or one form of peripheral data input/data display equipment. If one does then one risks falling prey to the "closed shop" syndrome - irreplaceable or non-upgradeable hardware and non-portable software.

Despite the misfortunes of the "fly by night" developers and distributors of VR and "quasi-VR" products, there are other companies who are becoming extremely active in the field of VR and will undoubtedly give SG a "run for their money" in the not too distant future. These include Kubota, with their DEC Alpha based platform and Evans & Sutherland with their Freedom graphics system range, not to mention the ESIG 2000 and subsequent platforms.

VR and telepresence: Nice application ... shame about the market

Advanced Robotics Research Limited (ARRL), based on the campus of Salford University in the North of England, was the first establishment in Europe to exploit this new technology as a means of achieving telepresence (Stone, 1992a; 1994). ARRL was formed in 1988 to operate a National Advanced Robotics Research Centre under the UK Government's Advanced Robotics Initiative, with a 5 year grant backed additionally by investment from a group of leading industrial companies, such as British Nuclear Fuels Limited (BNFL), Hunting Engineering Limited, and the National Nuclear Corporation.

From 1989, the ARRL's Virtual Reality & Telepresence program made significant strides towards providing the ultimate form of telepresence, principally for nuclear and subsea applications. Research at ARRL concentrated on the use of Virtual Reality (VR) hardware and software as tools for achieving telepresence, not only to provide the means of displaying and interacting with three dimensional computer graphics, but also to demonstrate the power of being able to convert sensory data (from a scanning laser rangefinder in the ARRL's work) and to control real robot manipulators and vehicles.

The first phase of the Company's work in this area reached a successful conclusion in September 1992, with a subsequent Telepresence and VR Demonstration staged at the House of Lords in November. There were five key elements to ARRL's achievements. In brief, they were:

Laser Imaging and World Modelling, based on an infrared scanning laser rangefinder, capable of producing extremely dense range information. The analysis of this information was concerned with processing multiple dense range images into a simplified (segmented), fused format. See also Stone (1992b, 1992c).

A Priori World Modelling from CAD Databases, converting existing data to exploit the power of VR for interactive "fly throughs", featuring (amongst other features) data/detail switching, multiple instancing and animated bill board texturing.

Mobile Vehicle and Manipulator Telepresence. In September of 1992, the first demonstration of the integrated VR/Telepresence system took place, with a remote human operator driving a Cybermotion K2A robot vehicle, equipped with a high speed, head slaved stereoscopic video system A standard VR headset was employed and reasonable stereoscopic video pictures permitted the operator to drive the vehicle over short distances safely and efficiently. Also featured in the demonstration was the use of VR peripherals - the same headset plus ARRL's Commander Spatial Controller - in the performance of inspection and handling tasks using a Puma 562 Robot Arm. The Puma teleoperation system had been enhanced using a brand new control system developed by ARRL, which allows full six degree of freedom operation without the traditional problems inherent in this type of robot, such as driving joints out of limits or losing degrees of freedom through axis alignments (singularities).

Virtual "Teamwork", feeding stereoscopic images from the ARRL's VR computer for 3D display on a special silvered screen, using a pair of liquid crystal video projectors with polarising filters. Regardless of whether the view was virtual in nature or generated from stereoscopic cameras mounted on the robots themselves, team observers experienced a full 3D image by wearing polarised filter spectacles, complimentary to those used on the projectors.

Nanopresence, converting data from a scanning tunnelling microscope (STM; see also Donnelly, Connell et al., 1993) into polygonal form with smooth shading, thus allowing an operator to "fly" over Angstrom level surfaces, where important features such as surface imperfections or implantation damage would become visible to the human eye.

Selling the "vision" of VR to industry

VR and its associated technology base is a very exciting and stimulating field in which to work or carry out research. But one question facing ARRL's research team was would VR ever find its way out of the arcade centres and research laboratories and into industry? Were there any applications which might warrant - technically and commercially - the introduction of a radically new form of human system interface? Certainly there are a wide range of potential applications being researched across the globe, from medical visualisation and molecular modelling to architectural "walk throughs", from advanced fighter cockpit and future infantryman research to virtual Olympic stadium design and skiing trainers. Yet, m the main, these applications are still confined to research laboratories. Even at Siggraph '93 and the later VR Exposition in New York (December, 1993), questions similar to those being asked over 3 years ago were still evident - superb graphics, excellent real time performance, but what are the applications?

Selling the vision and technical potential of VR to a sceptical and money conscious market can prove to be a highly challenging and sometimes frustrating exercise. The misleading promises surrounding VR, fuelled by the media and entertainment industry, have been, without doubt, the cause of most of the frustration, leading to a widespread perception that the state of the art in VR is very poor indeed. The effect of the VR "hype" was that, until recently, many potential industrial users openly declared that they would never contemplate procuring a "child's game" for serious commercial applications. Having raised the hopes of other, more enthusiastic sponsors, the misleading statements about what VR technology could already achieve produced serious disappointment once they had experienced the primitive graphics and display technology first hand.

Rolls Royce plc: VR for aero engine maintenance

The first important step in introducing VR into industry came about following contact with Rolls Royce plc of Derby, who approached ARRL with the aim of evaluating VR as a means of improving its design and maintenance evaluation procedures (see also Angus & Stone, 1993).

In April of 1992, the future potential for using VR technologies to compliment computer aided design activities (based on the CADDS4X Package) within Rolls Royce's Digital Pre-Assembly - DPA - Department was discussed, especially with regard to the problem of assessing maintenance issues. Physical full size replicas of aircraft engines (PPAs - Physical Pre-Assemblies) cost several million pounds to fabricate, yet it is during the fabrication process that some problems of maintainability become apparent. In particular there are those problems which may arise due to the limitations of conventional CAD workstations to provide their users with an "intuitive" or natural view of the engine and its components. VR was considered to offer a logical design step between the DPA CAD modelling activities and PPA fabrication phases. "Immersees" could experience the service layout of an engine and could provide an early assessment of the planned distribution of services - pipework, gearboxes, brackets, and so on - for their ease of maintainability.

The first stage in the feasibility study involved an investigation of the portability of the DPA CADDS4X files of the Trent 800 civil engine onto ARRL's main VR computing platform, at that time a Division SuperVision System. This was found not to be a straightforward exercise, due in part to the solid modelling nature of CADDS4X. Nevertheless, after a period of around 2 months, and using the CADDS4X stereo lithography processing option in the conversion process, the first port of the Trent engine casing and lower bifurcation pipe assembly was achieved in the form of simple flat shaded groupings of polygons. Recent work has identified a more efficient and robust CADDS data conversion path, which is being refined at the time of writing.

Due to the complex interpositions between elements of the converted Trent model, the pipework assembly was found to generate excellent stereoscopic images, even using commercial LCD based (low resolution) VR headsets. However, the polygonal nature of the model lacked a degree of professionalism, and interaction with the pipework - withdrawing individual elements using ARRL's Commander hand grip - was not possible. Over the ensuing months, a series of polygon optimisation programs were written which improved the rendering quality and "fly through" speed of models ported onto Supervision. These were applied to the Trent 800 model successfully. Additional routines were written to colour code individual pipe routes selectively and to segment pipes into groups of objects, thereby allowing detail switching and their removal and manipulation by engineers. The feasibility study culminated with a demonstration of the virtual Trent engine to senior Rolls Royce personnel, using both immersion and stereoscopic projection displays.

UK NIREX: The UK's first industrially funded VR contract for industry

Another important VR project which helped ARRL to sell the vision of VR is being carried out under contract to UK NIREX Limited of Harwell. Whilst specific technical details about the work cannot be released, for reasons of confidentiality, the project relates to the development of a virtual model of the Deep Waste Repository currently being considered for the safe and cost effective disposal underground of solid low and intermediate level radioactive wastes at Sellafield. The requirements envisaged for such a model are to permit investigations of VR technologies for such purposes as interactive design and visualisation and training procedures. These investigations would employ the basic UK NIREX Repository computer model as constructed using a variety of standard computer aided design and VR graphics rendering packages. An important aspect of this requirement is that changes to the graphical model and its components must be efficiently supported, to occur in line with the phased development of the Repository. Commencement of operations is scheduled for the next Century, with a subsequent operational period of 50 years.

The Project (see also Byte, September 1993, page 38) takes the form of a consultancy and modelling services agreement, with topic areas to include performing benchmark tests of candidate host computers (eg, the Silicon Graphics RealityEngine Series, the Sun Sparc10-Evans & Sutherland Freedom graphics boards, etc), developing CAD-VR conversion standards and providing draughtsperson training, and modelling the Sellafield area using processed data from Ordnance Survey Digital Terrain Data. UK NIREX wishes to be kept informed regularly of new developments in VR data display and input devices, so that the Company can choose the appropriate time to move credibly from a non-immersive to a semi (3D projection/BOOM), or even fully immersive capability.

Virtual reality meets their lord and ladyships

In the height of a massive security alert against terrorist attacks in London, a solitary van, laden with expensive robotics and Virtual Reality equipment made its way to the centre of British democracy - the House of Lords. The van arrived at its destination unhindered, unlike the author, who received a verbal reprimand from the London Constabulary for having fallen asleep in an underground car park, just a few yards from the Parliament buildings!

The presentations and demonstrations to their Lord and Ladyships occurred with flawless timing, lasting all of two hours. One of the highlights of the event was the rapid appearance and disappearance of one of the Lady Peers who, just prior to her departure was heard to remark, disappointingly, that she had been expecting a demonstration of advanced aerobics, not robotics! Another highlight for the VR Team was the attendance of Edward Simons, Chairman of the Allied Entertainments Group and one of the Executive Producers of Lawnmower Man, who did not appear to be very keen to travel to Salford in order to see what VR is really about (a vain attempt on the part of the author to try and inject a degree of "reality" into Lawnmower Man 2!).

A winter of discontent

As Christmas 1992 approached, with the DTI funds exhausted, the future of ARRL's VR Team was not looking very healthy. Many industrial visitors to the Centre continued to make the same frustrating comments, such as: "my son would love this in his bedroom" (see also Meckler's Virtual Reality Report, Volume 2(6), July/August, 1992). Undaunted, towards the end of the year, the Team identified two potential opportunities - events which would, it was hoped, provide a forum for lobbying the grant givers to inject much needed cash into UK industrial and academic VR efforts, not just for ARRL, but for British concerns in general (although, it must be confessed, acquiring some funding for ARRL was one of the author's hopes). These events were a VR Symposium in Warwick (in the Midlands), sponsored by the SERC, and the Christmas Lecture to the UK Department of Trade and Industry (DTI).

Coincidentally, ARRL was contacted in the Autumn of 1992 by one Elisabeth Geake, Technology Reporter of the New Scientist magazine. Geake wanted ARRL to provide a few comments to support a small feature covering Division Ltd's ProVision 100 launch, announced during a press day at their new premises in North Bristol. In passing, statements were made about the appalling lack of funding for VR research and development in the UK, especially as, in the main, companies and some academic institutions had helped nurture the country's European lead through investment of their own scarce funds. The Warwick and London events mentioned above were described with regard to the intention to deliver a few "home truths" to national research funding bodies. As with many press features, these comments were well and truly taken on board and the article appeared in New Scientist (12 December, No. 185 1; p. 18).

Stone is meeting with the Science and Engineering Research Council and the Department of Trade and Industry later this month and will tell them: "You back us up so that we keep our lead."
The SERC and DTI functions turned out to be none too useful from the point of view of obtaining a firm commitment to supporting VR technologies and applications. Excuses - now commonplace - were rife, such as "we're evaluating our options", or "we're not convinced yet that VR is mature enough to support as a technical field in its own right", or "we don't know what our position is ... we are undergoing a reorganisation, you know". Another, related problem was the DTI's "cooling off" attitude with technologies that were directly associated with, or appearing to evolve from anything to do with advanced robotics. Despite ARRL's achievements across the board in advanced robotics, there was a feeling that, because a UK Government Minister had not been photographed shaking the hand of an anthropomorphic robot (as might be orchestrated by the Japanese), advanced robotics endeavours in the UK had not been too successful. This is obviously a rather cynical view, but it does describe one of the ways in which technological fortunes in this country can be cemented or broken by Government attitudes.

What did turn out to be the turning point out of all this was the fact that the British Broadcasting Company (BBC), ever hungry for a controversy, read the New Scientist article and decided to send a film crew to Salford early in January, with the aim of running a feature on the 9 O'clock News. Another coincidence helped to make the feature a success. It just so happened that, at the time of filming, ARRL was in the process of evaluating a Silicon Graphics RealityEngine for possible future VR use (funds forthcoming, of course), as well for providing a better real time capability for the Company's simulation work, using Deneb Robotics' IGRIP Package.

In fact, the feature was not screened immediately after the filming session. Certain other world events, namely the latest war mongering efforts of one Saddham Hussein - were monopolising time on air. Nevertheless, and to the ultimate relief of ARRL staff (and their families), the piece finally went out on 19 January. The VR feature lasted for around 3.5 minutes and included an immersive fly through of the Rolls Royce Trent 800 Engine described earlier, the now familiar Performer Town demonstration hosted on the RealityEngine, simulated nuclear fuel rod recovery modelled using IGRIP, and some of the University of York's molecular modelling research for Glaxo. It also included a statement by Edward Leigh, now the ex-Minister for the DTI:

If it is apparent that there is a market failure in this industry, then we will consider investing in this further. Certainly our colleagues in the Science and Research Council are still investing in this project and so the Government is committed to it and is committed to making it work.
As an aside, shortly after the 9 O'clock News item, the BBC's new "virtual" studio was announced in the popular press, and on 13 April, viewers were exposed to the sight of a massive crystal crest (modelled using Vertigo) which rotated on screw to bring the Company's motto - Nation Shall Speak Unto Nation - into view. Virtual Reality? Not at all - each second of motion (ie, 25 frames) took around an hour to render!

After the VR feature, the response to ARRL from industries was incredible, to put it mildly. Many representatives expressed surprise that, for once, VR was not being portrayed as the ultimate in games and leisure technology, but was being used in genuine industrial applications, with real time graphics of a credible and professional quality. Furthermore, the subsequent response from the DTI also came as a bit of a surprise, suggesting that it might now be possible for ARRL to put together a proposal for funds from an information technology initiative they were currently coordinating in the field of Computer Supported Cooperative Working (CSCW). However, for reasons described later, this was not pursued.

As it happened, a British VR project did receive some funding from the DTI - Virtuosi - a collaborative program involving British Telecom, GEC and Division. The project was designed to investigate virtual office working across an ISDN network. However, when this was announced in the press, the picture painted was one of disappointment and disgruntlement on the part of the companies involved. Rather than extolling the study's virtual virtues, the feature focused on the fact that an originally suggested budget of 3 million had been reduced to 2 million and was to be shared among 7 other projects, leaving Virtuosi with only a fraction of the funds originally requested.

The Virtual Reality and Simulation (VRS) initiative

During the early stages of putting a possible collaborative team together for a project of some relevance within the CSCW Initiative, it became clear that many more industries than was first thought wanted to "get in on the act" and support some form of collaborative venture in the field of VR. With, perhaps, one exception, none of the companies ARRL was in discussion with had the resources to invest in setting up an in house "professional" VR development unit during the current recession. They did appear, however, to be convinced (to greater or lesser degrees) that their technical and commercial businesses would ultimately benefit from adopting VR. With the enthusiasm evident from industry for setting up a VR program, ARRL consequently decided to "go it alone" and not submit proposal for Government funding. With the administrative and bureaucratic problems such funding would inevitably bring, plus the Virtuosi experience described above, this turned out to be the right move.

The negotiation of terms and conditions of what was to become Europe's first and largest collaborative venture with companies with differing intellectual and legal requirements took a good five months. Nevertheless, on 25 May, 1993, the Lord Wade of Chorlton launched ARRL's industrial project under the general title of Virtual Reality and Simulation (VRS). The aim of VRS, initially planned as a two year program of work, is not only to keep British Industry abreast of significant international developments in the field, but also to demonstrate to participating companies the commercial value of VRS. At the end of the two year program, VRS will provide the participating companies with sufficient know how to introduce the technologies into their own businesses with minimal technical and financial risk.

VRS is the first European initiative of its kind fully funded by the industrial sector. Companies were invited to join the program on one of 4 tiers, depending on their status and perceived role: Full, Technology Watch, Small/Medium Enterprise (SME) or Associate Membership. Full Membership has been taken up by those companies who already have a well defined application and wish to sponsor a demonstration of the application using ARRL's VR and simulation resources. Technology Watch is a grade of membership designed to accommodate those companies who wish to keep a close watching brief on short term developments within VRS, prior to choosing an application of their own. The SME Grade of Membership was offered to two local companies involved in civil engineering and architectural practice. With areas of VR applications interest stated in parentheses (where known), the initial list of VRS Full, Technology Watch and SME members included:

Another important feature of VRS is the presence of Associate Members. An Associate Member is a collaborator who, through an association with the field of VR and simulation (ie, having access, through development or distribution, to relevant data, hardware and software products), adds value to the Initiative by contributing such data and products either for a discounted fee or free of charge for use by ARRL as deemed appropriate in specific VRS Projects.

Although the PR value of involvement is an important benefit of collaboration, Associate Members stand to gain primarily through the exposure of their contribution to others via the Projects undertaken by ARRL for Full and Technology Watch Members of VRS. One of the aims of the Initiative is to provide collaborators with sufficient knowledge and relevant code to implement Virtual Reality and Simulation within their own businesses. Therefore, the sales potential for Associate Members' products and/or services is significantly enhanced through involvement with VRS.

Associate Members, particularly those involved in the development of VR and simulation hardware and software, will also benefit from feedback from ARRL's specialist personnel on the performance and limitations of their products, thus enhancing future releases. Current Associate Members include:

The future?

One encouraging, recent development, is the signing up of a British company whose roots are, to many outsiders, well established in traditional retailing communities. The Cooperative Wholesale Society, with whom ARRL has had a relationship for over a year, joined VRS as a full member in October, 1993.

Cooperative Wholesale Society (CWS)

Founded in 1844, the CWS is a major UK business with a turnover of some 3.25 billion, excluding its activities in banking and insurance. The CWS is a leading retailer of both food and non-food products, a travel agent, and a bank and insurance company. It also has a substantial manufacturing base. For many years the Cooperative Movement has been viewed as innovative, boasting a number of technology firsts in banking and retailing. As the latest member of the VRS initiative (at the time of writing), the CWS views its involvement as an essential component of its continuing commitment to remain at the forefront of new technology developments.

The design of supermarkets and the planning of space within them is a time consuming and costly process. Every product on display within a supermarket is there to be purchased. The positioning, volume, lighting and advertising of each and every product is designed to achieve maximum sales potential. Currently, the visualisation of product layouts is performed using planograms. At their best, these are two dimensional photographic images. At their worst, they are two dimensional drawings showing coloured boxes. VR is considered as having the potential to provide the manager with a powerful tool, allowing him to visualise data in an easy to understand manner. VR is seen by the CWS as an exciting new way to visualise shelf layouts and to allow marketeers to interact with space planners, product buyers and supermarket managers in a way never before imagined.

Developments in electronic point of sale systems, electronic shelf edge pricing, customer tracking systems and intelligent packaging can all be combined with VR modelling using intelligent computing techniques, to enable the retailer to design, model and predict the performance of products, staff and consumers. At a time when retail margins are tight and profits are hard to come by, the use of technologies such as VR can be seen as key to the continuing profitability and success of a business, such as the CWS.

The local academic connection

The aim of a recently established formal link between ARRL and the nearby University of Salford concerns the development of a National Centre of international significance in the development of virtual environments for construction. To this end, the author was recently appointed as Visiting Professor within the University's Department of Surveying (with effect from 1 May). This post involves a nominal one day per week effort, assisting academic staff to realise the National Centre. At the time of writing, Sense8 WTK and Dimension VRT systems have been installed and efforts are working towards the linking of these systems with knowledge based architectural and surveying tools, to provide an intuitive front end for educational and research opportunities.

The ARRL Team, as part of its commitment to the local Universities' Federal Multimedia Agreement mentioned above, has also been assisting University College Salford in the establishment of a new BSc Honours Course in Video, Imaging and Communications Technology. The two course modules being coordinated by ARRL personnel are:

Human Factors (Year 2), covering current trends in the emergence of National, European and International standards in Human Factors and Human Computer Interaction. It was felt that the inclusion of a basic grounding in Applied Psychology and Human Factors is essential in a course of this nature, equipping the student with an appreciation of the importance of considering the characteristics and limitations of human beings as information processors in a system design or procurement process (including VR).

Virtual Reality (Year 3), emphasising that VR is fast becoming the defacto standard for interfacing the human operator with advanced computing applications involving the manipulation of 3D, computer generated imagery. Again, a course of this nature was deemed to be incomplete without a thorough grounding in VR technologies and practice.

The UCS Degree Course successfully passed a validation review on 27 May 1993 and is due to start in October (even before the validation, over 30 potential students had applied).

The Commission of European Communities (CEC)

Europe is ramping up to launch a major effort in VR. The Commission of European Communities (CEC) has already placed small consultative contracts with a group of key personalities within the Continent to advise the Brussels Coordinators as to developments and trends across the globe, including reviews of hardware, software and applications. This effort will pick up sometime early in 1994 with the likely establishment of a European Advisory Committee in VR, involving representatives from many countries. Within 18 months to 2 years, and if all goes according to plan, the scene will be set for an exciting era in the evolution of VR in this part of the World.

Don't theorise about it, try it!

If a newcomer to the field of VR happens to acquire a copy of The Virtual Reality Primer (Larijani, 1993), he might be forgiven for assuming that, in Europe, the UK had very little effort under way in VR and that the country was not considered worthy of an entry under a section entitled "Other Major Players: Germany, France & Japan", in Chapter 15. If the author and editor had performed their research with just a little diligence, then they might have discovered that, not only does the UK lead Europe in its VR endeavours, but the so called other major players openly admit this fact as well!

The launch of ARRL's VRS Project has demonstrated that the "child's game" attitude to VR and associated technologies is disappearing fast. VRS has demonstrated that industry now wants to be involved in VR - over a matter of a few months it has gained credibility, at least from a non-immersive standpoint. It is the Author's opinion that, with technical improvements gaining pace, it will not be long before full immersion for selected applications will also become the accepted norm. Hopefully, with national and international central funding cutbacks, VRS will not be the first and last Initiative of its sort. Only through exposing industries throughout the World to the potential of Virtual Reality will the field expand into a technologically acceptable means of designing and implementing human systems interfaces in years to come.

As Bob Jacobson stated during a 1992 Singapore based conference on multimedia, the UK's VR effort is characterised by applications focus and a good "national" sense of humour. ARRL's initiative is testament to this. VRS is not just the result of intensive commercial lobbying of British Industry. It is the result of a proven track record in making VR work, and of the enthusiasm, good humour and beliefs of the researchers involved, despite the frustration and worry caused by a deep recession. Where VR at ARRL, (or Salford in general) goes next, whether it is in the form of rolling VRS membership, individual contract research studies, or in an integrated commercial academic form, is hard to say. VRS is being scrutinised by many outsiders and the ARRL Team is conscious that the Initiative has to work, not only for the benefit of the Company, but for the accelerated acceptance and uptake of VR in general. Whatever happens, it is the genuine hope of the team that Salford can provide a true focus for industrial and academic VR efforts in the United Kingdom. It is also the genuine hope that this paper will encourage VR project champions in Australia to rally, spread the word and ensure that the technology really does graduate from the infantile status of arcade escapism to one characterised by a mature, leading edge, technological and commercial professionalism.

References

Angus, J. & Stone, R. J. (1993). A focused approach to the assessment of virtual reality as a design tool in Rolls Royce. In Proceedings of The VR User Show, Conference and Exhibition; London; Dec.

Anon. (1991). Virtual reality: The next revolution in computer/human interface. Michigan, USA: Matrix Information Services.

Donnelly, S. E., Connell, A. P. et al. (1993). A scanning tunnelling microscope for the study of surfaces irradiated with low energy ions. Journal of Vacuum Sciences and Technology, March/April.

Fisher, S., et al. (1988). Virtual interface environment workstations. Proceedings of the Human Factors Society 32nd Annual Meeting,

Krueger, M. W. (1991). Artificial reality II. Addison Wesley.

Larijani, L. C. (1993). The virtual reality primer. McGraw-Hill Series on Visual Technology.

Pimental, K. & Teixeira K. (1992). Virtual reality: Through the new looking glass. Intel/Windcrest/McGraw-Hill.

Rheingold, H. (1991). Virtual reality. UK: Secker & Warburg.

Stone, R. J. (1992a). Virtual reality: A tool for telepresence and human factors research. In Proceedings of Virtual Reality Systems (International State of the Art Seminar). British Computer Society, London. Also published in Earnshaw et al. (Eds.), Virtual reality systems. Academic Press, 1993 (Chapter 13).

Stone, R. J. (1992b). Towards telepresence: Imaging's answer to virtual reality. Advanced Imaging (Special Issue), August.

Stone, R. J. (1992c). Virtual reality and telepresence. Robotica, 10.

Stone, R. J. (1994). A day in the life of British virtual reality: Will the UK's answer to Al Gore please stand up?!! Virtual Reality World, January/February.

Sutherland, I. E. (1964). The ultimate display. Proceedings of the IFIP Congress, 506-50.

Endnotes

  1. As featured in the Sun - a UK tabloid newspaper - published on 1 April, 1993!

  2. Note that the term Virtuality has, since the Company's UK flotation in 1993, become the new trading name of W Industries.

Author: Professor Robert J. Stone
Advanced Robotics Research Limited, University Road, Salford M5 4PP, UK
Tel. +44 06 1745 7384 Fax. +44 06 1745 8264

Please cite as: Stone, R. J. (1994). Virtual reality comes of age. In C. McBeath and R. Atkinson (Eds), Proceedings of the Second International Interactive Multimedia Symposium, 608-618. Perth, Western Australia, 23-28 January. Promaco Conventions. http://www.aset.org.au/confs/iims/1994/qz/stone.html


[ IIMS 94 contents ] [ IIMS Main ] [ ASET home ]
This URL: http://www.aset.org.au/confs/iims/1994/qz/stone.html
© 1994 Promaco Conventions. Reproduced by permission. Last revision: 20 Jan 2004. Editor: Roger Atkinson
Previous URL 4 Oct 2000 to 30 Sep 2002: http://cleo.murdoch.edu.au/gen/aset/confs/iims/94/qz/stone.html