Few companies turn away business, but Intelligent System Solutions Ltd does it all the time. The Salford-based company is in the business of virtual reality and together with the University of Salford develops hardware and software for virtual applications. It also assesses the efficacy of existing products and believes virtual reality should improve a product or service, not just be added to make dull things seem exciting. And holding those views, managing director Professor Bob Stone says we will turn people away if the application they propose doesn’t warrant the use of virtual reality. At the moment the company is working to determine the real use of virtual reality. It is undertaking a controlled psychological study to see how virtual reality affects and potentially improves memory recall. One group of strangers to the University of Salford will be shown a virtual model of the campus and then given a map and told to find a particular building.
Virtual gloves
Another group will have to navigate the campus but will not get to see the virtual model. The two groups’ performances will be studied to see how they differ and, more importantly, why. But this is not just blue sky research: Intelligent Systems’ work may be used in work for ICI Plc to train petroleum plant employees on the layout of sites. Stone’s background is in psychology and ergonomics and this has strongly influenced the company’s philosophy. He believes the company’s pragmatic approach to virtual reality is the basis of its success. We are expanding and doing exceptionally well. And that expansion is by word of mouth, he added. In the past year the company has evaluated the use of virtual reality systems for Rolls Royce Plc where it has a ssessed how useful such systems would be in an application to determine how maintainable a new engine design is. The results backed up what the company already thought. The [hardware] technology has got a long way to go, said Stone. The user needs to be able effectively to navigate himself around an engine to repair it. The virtual reality system helps decide in advance how maintainable the engine is because it can be ‘virtually’ repaired before it is actually built. The number one problem was reliability. For the Rolls Royce application, pointing with a virtual reality glove to where you want to go is not sufficiently like real life. It is impractical for deciding how easy a particular engine would be to maintain. One reason for this is the inaccurate recognition of movement using virtual gloves. Users had problems repeating actions using virtual gloves. Often when they thought they were carrying out the same instruction, the glove instructed the computer to do something completely different. Intelligent Systems has taken a whole series of headsets and evaluated them for Rolls Royce. Nausea and disorientation were experienced by many research participants using virtual reality headsets. The headset fools the body into thinking it is moving and to avoid nausea, something was needed to ground the user into reality. So a wooden platform was made for the person to stand on. The headset was fitted with a graphical control panel. The virtual hand presses the graphical buttons and gives the impression the platform is moving backwards, forwards and to the sides, thanks to the graphical representation of the platform in the headset. The wooden platform does not really move but the virtual one does. The users didn’t get disoriented and this system was preferred by all the people that tried it because they had something real to hold on to if they felt unstable.
By Abigail Waraker
Users also felt they had greater control over what they were doing this way. Something else that left the users feeling lost was the sense of free flight through a scene when using a headset and glove. In particular, overshooting the target. With the glove you point in the direction you want and off you go. Unfortunately it is very easy to miss the mark and find that where you intended to go ends up behind you
. Users were finding themselves flying through or past the point they were heading for and lacked the control to stop where they wanted. The research found that the more human movement can be constrained, the more secure the user feels. Unconstrained flight is not on. said Stone. The Rolls Royce application requires controlled movements. Moving uncontrollably through the engine is not applicable to this specific task. Stone accepts that this is very much an issue for the person writing the software. The programmer can write in constraints on the amount of freedom the user has, but with the wooden frame, which can be seen in the virtual world as well, the user has a greater sense of where he actually is, as he has a point of reference and so doesn’t overshoot the target so much. Stone compares the system to a bungee jump. It would far easier to put out a fire on the end of a fire service crane than if you were on the end of a bungee rope. Unconstrained movement makes carrying out a task much more difficult for the user. The two main problems with headsets are the field of vision and the image quality. If the headset is liquid crystal display-based, it has wide-angled optics. To generate a sense of presence, this wider field of view is better, but the image quality is appalling, according to Stone. The other type of headset, the cathode ray tube-based variety, has great image quality, but it’s like looking through binoculars. So the reverse problem is true. The image quality is good, but the sense of actually being in the scene isn’t. Intelligent Systems believes it would not be difficult for a company to develop a system with both a wide field of vision and television quality graphics or better, preferably with a ú5,000 tag. And this is the challenge he puts to the hardware community. Headsets are overpriced for what you get anyway, he added. Stone admitted some headsets are not bad, but exactly which ones is the subject of an internal report for Rolls Royce and will remain secret. Intelligent Systems has no intention of entering the market to make headsets, though. What interests it is building handsets.
Hand controller
It estimates that a typical virtual reality glove with tracking costs around ú17,000, but a hand controller with tracking costs around ú5,000. And 75% of the ú5,000 is the cost of the tracker. In light of all the problems it has faced with gloves, one of its next steps is to look at replacing them. Pressing buttons has proved to be a more reliable method of control than moving fingers linked to sensors in a glove. Each virtual application requires a new feature from a hand controller, so it is looking to make one that can be changed to suit the needs of different types of applications that will arise in the future. Although, in the virtual reality market as it stands, Intelligent Systems says handsets form the most reliable part of a virtual reality system and they are relatively cheap. Intelligent Systems’ success seems rely on its continued questioning and testing of the current equipment on the market and on companies coming to it for advice on the systems that are available. It is not interested in applying virtual reality as a gimmick, but as a way of really enhancing an application. Intelligent Systems drools when faced with new virtual reality hiccups. We are now breaking into a new area for ourselves. We are working on a virtual reality maze for a television games show and it’s brought up some interesting technical problems.
