Interactive design methods in virtual reality with haptics.

摘要

This research demonstrated the use of virtual reality and related interaction technology, such as haptics, in an immersive virtual design application. The goal was to present an effective methodology for interactive design where engineers alter products and investigate updated analysis results in real-time using a virtual environment. Research was presented as a series of four papers.; Chapter 2 covered a history of the interactive design methodology and it's applications followed by the largest hurdles to effective interaction. Solutions where presented and their effectiveness demonstrated.; The third chapter examined some limitations of the custom mesh-free analysis method used on the application. A search was performed for suitable replacement software with an emphasis on Open Source applications. The Tahoe program was selected, tested against the custom implementation, and integrated with the application.; In Chapter 4 the calculation of shape design sensitivities for stress approximation were examined. Limitations of the existing finite differences technique were explained and a search of available techniques was made. After considering discrete derivatives, continuum derivatives, and automatic differentiation, and discrete derivatives approach with exact numerical differentiation was selected, implemented, and tested. Results indicate the discrete derivatives approach to be a fine replacement.; Chapter 5 considered the use of haptic or force feedback as an additional channel of information for the designer. Several haptic devices and their uses were examined, and a PHANTOM 3.0 was selected for use with the immersive design application. Networked haptic feedback was added to the application and several techniques for mapping model stress state to device forces were investigated. Finally, a simple pilot study was conducted to help determine which feedback styles were considered most useful to a potential designer.; The area with the greatest potential for future work would seem to be the use of haptics as a design tool in the application. A proper user study should be conducted to sort out which stress-force mapping techniques are the most effective for a designer. One should also try to determine just how useful the additional haptic feedback generally is when working in the virtual environment.; A second area worth investigating further is the use of automatic differentiation routines for the computation of shape design sensitivities. Throughout the course of this research automatic differentiation has made many improvements. With programs as large as 400,000 lines and more now being completely differentiated, it appears to be fast becoming a practical analysis option. This would also make implementing additional types of sensitivity analyses much more straightforward, and would be a powerful addition to the methodologies of interactive design in virtual reality.