The objective of this award is to develop a mathematical theory and computational framework for the automatic generation of context-dependent simplified models to support interactive virtual assembly applications. Interactive virtual assembly is emerging as an important tool for evaluating ease-of assembly of proposed products, and training assembly operators. With the advent of low cost personal virtual environments, interactive virtual assembly holds the promise of replacing expensive, and time consuming, physical prototyping and training. However, real-time interactions in low-cost virtual environments can only be achieved through judicious model simplification. Through a fundamental understanding of the interplay between model simplification, computational speed and accuracy, a framework will be developed for optimizing model simplification for virtual assembly.
If successful, the results of this research will enable interactive virtual assembly with provable performance guarantee on low cost personal virtual environment, leading to a significant increase in the use of virtual assembly technology in design and training applications. The proposed work is expected to lead to: (1) reduction in training cost by eliminating one-on-one training, and reducing the use of physical prototypes, and (2) better assembly designs by ensuring ease-of-assembly. Efficient and effective training methods will lead to a more agile workforce that is capable of quickly adapting to the changing requirements of the manufacturing industry. In addition, the planned research and education integration activities and outreach activities will: (1) familiarize graduate, undergraduate, and high school students with the use of virtual assembly and model simplification technologies, and (2) increase awareness among industry about potential usage of virtual assembly on low cost personal virtual environments.
