Like many emerging technologies, virtual reality (VR) has overcome the excitement and hyperbole of its early promise and matured into useful product development tool in a variety of manufacturing industries. Though subject to broad interpretation, VR applications generally include interactive frames per second), user-driven stereo graphics, controlled with some form of physical tracking of the user, to produce a perception of immersion within an artificial environment. Display devices range from desktop monitors, to head-mounted displays, to large-screen panels or tables, to room-sized surround-screen environments.
With roots primarily in government and university research laboratories, VR enjoyed a flurry of commercialization in the early- to mid-1990’s resulting in a numerous hardware and software technology providers. However, the relatively high cost, lack of compelling applications, and lack of integration with existing processes, conspired to dampen the initial excitement of many manufacturers for VR. As a result, the past ten years has seen a dramatic consolidation in the commercial VR technology industry. Like many other potential users, manufacturers took a slow, but steady approach to incorporating this new technology into their product development processes.
During the same period, research at universities, government laboratories and companies has continued to push the VR envelope. Progress has been accelerated substantially by two over-arching technological trends. First, the VR research community has, to a large extent, embraced the open source model of software development. Large and active developer communities have evolved in support of a wide variety of enabling technologies including high-performance scene graphs (e.g., www.opensg.org and www.openscenegraph.org), VR application frameworks (e.g., www.vrjuggler.org and http://diverse.sourceforge.net), and even interactive physics engines (http://opende.sourceforge.net). Open source not only leverages the mutual productivity of the research community, but also facilitates the transition of useful applications from research labs to industrial end-users, or even to general commercialization. Second, driven primarily by the computer games market, high performance graphics hardware has matured to the point at which clusters of commodity hardware routinely out performs dedicated graphics supercomputers of just a few years ago. The implication of this dynamic is profound–it makes VR technology economically feasible for a broader array of potential applications.
VR in general is a broad interdisciplinary field spanning a wide spectrum of technical topics, as well as the psychology, communication, and social sciences. Although important research continues in technologies that enable VR such as graphics hardware and software, display devices, human interface techniques, and tracking, to name a few, this issue of JCISE focuses on the potential use of VR technology to facilitate all phases of the product development process. It presents a snap-shot of the current state of the VR art as applied to industrial product development.
Despite this relatively narrow focus, research challenges in this area are rich and numerous. Potential applications of VR in product development span the entire spectrum of activities including finance, marketing, design, engineering analysis, manufacturing, process planning, service, and recycling. VR excels in applications characterized by complex relationships among product data or between data and product models that benefit from human interpretation and decision making. For example, applications that enable exploration of patterns or trends in abstract high-dimensional data are being explored for applications as diverse as finance, marketing and product analysis data.
One of the most pressing issues facing industry is data integration. The (CAD) systems used to author product models are generally not suited to producing a representation conducive to large scale, frame-rate guaranteed visualization required by VR applications, e.g., a scene graph representation. Although addressed to some degree by commercial providers of visualization software (such as UGS PLM Solutions and Dassault Systems) there is no general non-proprietary way to convert a CAD assembly into a representation suitable for VR. Incorporating persistent pre-existing texture maps in these high performance models is an additional challenge. Similarly, persistent predefined kinematic relationships or dynamic behavior among parts in a scene graph (or alternatively authoring them in the VR environment) is an open challenge. The open source scene graph efforts mentioned above represent a promising basis for this research, but as yet, it remains a major bottleneck to further industrial adoption of VR.
Another important trend is the incorporation of VR as an integral user interface for sophisticated engineering analyses. Rather than treating the immersive environment as a post-process, visual interrogation tool for CFD, FEA, or other computationally intensive analyses, proponents of “virtual engineering” are seeking to integrate new user interfaces that enable modification of analysis parameters, or even prototype geometry, and compute results in real time using either approximation techniques or tightly integrated massive computational resources. By fusing the immersive aspect of VR for exploiting human interpretation of complex relationships with contemporary engineering analyses, this research has the potential to substantially improve the quality of analyses and accelerate the product development process.
Some of the most powerful applications of VR are those in which the scale of the subject product or data is comparable to that of the user. Human-scale design applications abound in automotive, aerospace and heavy equipment industries. Other applications are motivated by manufacturing processes that are human-centric, such as routing flexible hosing through a complex assembly. These applications drive a number of current research challenges including the incorporation of measured anthropomorphic data in immersive design for vehicle or device operators, the representation and behavior of virtual humans, incorporation of real-time, “human-in-the-loop” simulations, and software and interface infrastructures to support collaborative VR environments. These human-centric research efforts promise to extend the utility and value of VR into many new product development challenges.
As demonstrated by the diversity of the papers in this special issue, VR is a vibrant research topic with the potential for dramatic impact on a wide variety of product development challenges. The editor thanks the authors and reviewers who contributed to this special issue and hopes that it inspires continued research activity in this exciting field.