Air Domain Autonomous Control System Architecture (ADACSA)
Abstract
The proposed project, Air Domain Autonomous Control System Architecture (ADACSA), is designed to address a major problem with military airspace management. Since World War I the combat airspace has grown increasingly busier and more complex, with voluminous numbers of aircraft and other entities, such as unmanned aircraft, rockets, missiles, artillery shells, and mortar rounds. Current airspace management is chronically insufficient, even with improvements in computer hardware, software, and display technology that is now available to support airspace operations planning and management. Significant weaknesses continue to exist in the processes for airspace management and deconfliction. There are problems with the dissemination of airspace information, aggravated by battle tempo, which have caused fratricidal incidents and deaths. The airspace is now changing rapidly and requiring more dynamic, realtime airspace management and deconfliction. Changes include: an exponential increase in the deployment of UAVs, which will soon become autonomous and not remotely operated or supervised, and which will be deployed in vast swarms; and an increasing tempo in air and ground combat because of new technology and systems of systems incorporating distributed sensors, processors, weapons, and communications. New technology can provide faster, more accurate airspace management and control able to accommodate large numbers of entities in the airspace, including swarms of autonomous UAVs. We propose that a modified version of the autonomous intelligent control system architecture, developed by the Intelligent Systems Division (ISD) of the National Institute of Standards and Technology (NIST), known as the 4D/RCS, serve as a reference model architecture for the management and control of military airspace operations. The 4D/RCS reference model architecture was originally developed to serve as a framework for the autonomous intelligent control (with a human machine interface (HMI)) of an autonomous factory. It was then, over the decades, instantiated in individual and multiple autonomous vehicles, and other robots, whether in air, on land, or in water. The architecture can accommodate various artificial intelligence (AI) tools and techniques, including sensor processing, expert systems, neural networks and other machine learning techniques, and genetic algorithms. Its design and existing code for various modules were developed by NIST, an agency of the U.S. government, and thus are freely available for modification. An ADACSA based on the framework of this reference model architecture will solve critical airspace management and control problems and satisfy the functional requirements for military airspace management and control in the coming age of autonomous aircraft and swarms. After modifying the design of the NIST 4D/RCS reference model architecture to best accommodate the functional requirements and concept of operations of ADACSA, we will develop a ConOps for ADACSA and describe how ADACSA airspace management and control would work operationally. We will describe the human interaction with the system, including managers, users, and other stakeholders, and perform a risk and uncertainty analysis for the prospective ADACSA. We will determine how the prospective ADACSA will interact with the distributed operation of autonomous heterogeneous swarms of (relatively) low-cost UAVs and UGVs. We will simulate, model, and demonstrate ADACSA for example ConOps, and provide documentation: monthly progress reports, a final report, and a presentation.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Sep 11, 2020
- Source ID
- N000142012792
Entities
People
- Robert Finkelstein
Organizations
- Office of Naval Research
- United States Navy