Attribute-based multidisciplinary optimization of undersea vehicles

In an existing environment for designing undersea exploratory vehicles, the customer sets target values for certain attributes such as Range, Speed, Power, etc. and then desires a virtual prototype of the vehicle which includes variables such as dimensions and parameters associated with propulsion, guidance and control, hydraulics and other sub-systems. After a design is determined from the attribute specifications, simulation programs are used to estimate the performance and associated cost. However, while subsystem design servers exist, there is no coordination procedure among these various servers leading to a significant lack of automation. There is a need for a coordination/optimization strategy while maintaining the attribute driven environment currently used, the design process is currently done manually owing to the absence of a strategy to coordinate the various design servers. A methodology for automated design synthesis is presented here, using Collaborative Optimization. An example problem is presented. Non-gradient methods are used to circumvent non-smoothness of the problem and that of the collaborative optimization strategy. Future directions are identified.