The research objective of this GOALI award is to develop and evaluate methods to support natural human interaction with digital CAD models with a focus on simulating manual assembly tasks in an immersive virtual environment. The motivation is to develop tools that will allow engineers to mimic assembly of products while the products are still in CAD form in order to identify potential issues in assembly processes that could be alleviated through changes to the product design. Issues that arise on the shop floor are very expensive to address if they involve changes to the product shape, function, tooling, etc. Therefore, identifying and fixing these issues before the product is manufactured holds the potential for significant cost savings. The underlying research fundamentals involve simulation of natural human interaction with products in an immersive computing environment.

Intellectual merit

A large number of bench assembly operations are performed with two hands; therefore, the virtual assembly environment must also support bimanual assembly. Engineers in industry who are interested in exploring the potential use of virtual reality for assembly methods prototyping are uncertain about what device configuration is sufficient to obtain reliable feedback from participants in decision making tasks. One of our studies compared the use of two haptic devices to the use of a non-haptic glove on one hand with a haptic device at the other hand. This last configuration would be less costly for industry and provide the user with a larger workspace since the gloved hand can reach quite a bit farther than the small desktop haptic devices. We learned from this study the glove/haptic device configuration does hold promise for providing at least the same performance in virtual assembly as a dual haptic device configuration. Another interesting result was that half of the participants in the study chose the non-haptic glove and haptic device as their preferred interaction configuration. These are encouraging results that will lead to further research on the use of a glove/haptic device hardware configuration for virtual assembly.

Another one of our studies examined the effectiveness of virtual training compared to training with physical objects. Again, we are interested in the benefits of immersive computing technology to reduce costs to product designers/manufacturers. The results of that study indicated that although the participants in the virtual training took longer to complete the training, their performance on the real assembly task two weeks after training was improved, while the participants that were trained using the real physical objects demonstrated decreased performance after two weeks. This could indicate that users trained in the virtual environment retain the knowledge gained from training better than those trained with traditional methods. This result is consistent with a body of research in psychology that indicates increased training retention when training tasks require a lot of attention and concentration by the users.  

Our collaborators at John Deere hosted our visit to one of their manufacturing plants. We were able to observe a product review session that included engineers and shop employees discussing an upcoming product configuration change that would have an impact on the production process.  It was very helpful to hear the discussion between the two groups of people. Afterwards were had a better understanding of how our research could be used to facilitate such a review session. We were also able to go out in the factory and observe shop employees assembling products. The complexity of the process, including workstation layout, tooling, material handling, scheduling, etc. was a rich environment for study. These meetings helped us understand the specific needs of industry so that we could prioritize our research to...

Please report errors in award information by writing to: awardsearch@nsf.gov.