High Power Blue Laser Design by LaserAID, a Machine Learning, Artificial Intelligence Laser Design Tool

This Navy topic calls for the development of an improved method of designing blue emitting laser systems and the fabrication of a laser against performance requirements both technical and environmental.  Q-Peak, a division of Physical Sciences Inc proposes to design a hybrid fiber & solid state amplifier chain followed by non-linear conversion to generate output at a wavelength of 47x nm.  This laser needs also to be designed to be compliant with full military specifications for airborne lasers systems.  This is inclusive of the laser seed, amplifiers, and non-linear optics as well as the electronic drives and controls, and any required support systems for thermodynamics.  Complex laser designs consume a large amount of engineering resources, multiple iterations and prototypes to demonstrate required performance.  Project lifetime is long (~2-3 yrs) with considerable financial resources.  Reducing design lifecycle and attendant cost is major goal of this program and will be addressed by creating a neural network/machine learning software package capable of emulating the design process.  Codifying the knowledge base of the laser engineers and converging on an optimized design early will reduce the necessary engineering labor, produce a more Key Performance Parameter-compliant design and yield faster program execution at lower embedded cost.