Multiresolution Algorithms for Virtual Prototyping of Massive CAD Models

Principal Investigator: Dinesh Manocha  
Funding Agency: National Science Foundation 
Agency Number: DMI-0400134

Abstract
Virtual prototyping is often used to reduce design time, lower production cost, and improve the level of innovation in developing mechanical parts of varying scale: from nanometer-sized objects including nanoscale robots, storage, devices and nanowheels, to large man-made computer-aided design (CAD) structures including airplanes, powerplants, and submarines composed of millions of, parts. To help fully realize its potential, the proposed, research targets the solution of several well-defined CAD/CAM problems; including collision detection dynamic simulation and interactive display The research goals include: Algorithms: Design of new algorithms, and multiresolution representations for-fast collision and distance computations between polygonal and higher order models, penetration depth computation, dynamic simulation of deformable models and, interactive, display of massive geometric models. Software Systems: Development of libraries for interactive collision detection, dynamic simulation, and real-time modeling of massive datasets. These libraries will be distributed via the WWW along with the, software for interaction with massive datasets. Applications: The algorithms will also be used for virtual prototyping of large CAD structures with millions of parts, including models of the DoubleEagle Tanker and Boeing 777. The overall system will be used for maintainability studies and accessibility analysis, part-removal, and generation of 3D maintenance instructions.

Intellectual Merit: This research is expected to lay the scientific foundation for an emerging paradigm of physically-based modeling within large simulated environments. It includes developing new algorithmic insights and integrating them into a challenging application. The underlying representations, algorithms, and software systems for fast contact computation, interactive modeling of flexible objects, and motion planning will offer fundamental advances for robotics, CAD/CAM and simulation.

Broader Impact: The proposed research could lead to wide application of physically-based modeling and motion planning algorithms for virtual prototyping. Furthermore, the proposed application has the potential of making a significant impact on its own. The development of electronic maintenance instructions, parts removal for virtual prototyping, and interaction with large synthetic environments could easily result in annual savings of tens of millions of dollars for the CAD industry.