Publication

The goal of the Intelligent Armament Systems project is to advance defense systems capabilities through the development and application of artificial intelligence, machine learning, deep learning, and intelligent cooperation. In this project we developed algorithms, and designed and implemented real time experiments to demonstrate their technical feasibility. In addition to solving defense problems, our algorithms can also be used in intelligent systems to improve urban operations for first responders. Our results predict that our continued research will incorporate one or more technologies that will have a broad and profound impact on Weapons and Munition Armaments in the 21st Century and beyond.

Intelligent Armament Systems’ underlying technologies of artificial intelligence, machine learning, and deep learning closely align with the research initiatives at CCDC, and the future Digital Engineering Strategy of the Office of the Secretary of Defense (OSD).

Consistent with these initiatives, our ultimate objective is to create smarter armament systems, transforming complex armaments into systems capable of collecting and responding to internal and external situational data. The intelligence added to a defense system extends beyond simply adding software programs. Hardware, sensors, monitors, actuators and other components can also support the realization of greater system intelligence. Qualitatively, these additional components can serve as the senses (sight, hearing, smell, etc.) of the system. They can also serve as the actuators (speaking, touching, moving, etc.) allowing the system to respond to its environment. Taken together, these various enhancements comprise an integrated set of hardware/software components embedded within the larger scale system to create a smarter armament system.

Rapidly advancing physical technologies will continue to be main drivers of more powerful systems. However, integration of innovative intelligence through machine learning, deep learning and artificial intelligence will be an equal partner in these advances. This is because today’s systems are high performance, agile, and complex.

Indeed, a critical performance issue relates to real-time operation, namely the requirement that the intelligent add-on may help the armament system respond within the time intervals dictated by the environment in which it is operating. Future work will address the important issue of critical fast real-time operations using appropriate hardware and software. The emphasis throughout will be on embedding intelligence within electronic armament systems without adding substantial cost or sacrificing performance (speed, reliability, power dissipation, etc.).

An important research theme is the concept of design for intelligence, namely designing largescale systems that explicitly support system extensions using machine and deep learning algorithms. Future research programs will explore innovative opportunities for extending current systems to future systems. The objective of intelligent armament systems is not the result of disruptive technologies but rather the evolutionary addition of innovations, allowed by the increasing complexity of microelectronic components, software and learning algorithms, leading to new systems and new products.

Innovations embedded in existing systems can have a highly disruptive impact on their longer term evolution. Intelligent add-on capability to an existing, well-defined system can lead to integration of that same capability into later generations of the larger scale systems. This can profoundly change the future evolution of that system.