TECHNOLOGY AREA(S): Info Systems
OBJECTIVE: Develop a simulation architecture framework that enables development teams to integrate algorithmic models into a simulation solely through a collaboration layer/interface and software controls (requiring no modification to the simulation’s central architecture/engine software).
DESCRIPTION: This topic seeks a capability to easily integrate algorithmic models (e.g. gravity model, radar detection model) directly into a simulation architecture as modules with the objective of allowing distributed development of models by expert teams, while retaining the advantages of an integrated simulation (consistency of models, consistent input data formats, centralized data management, etc.). Because different models require access to the central engine software and pass different data items, the data items handled by the simulation architecture and its collaboration interface must be easily expandable without modification of the individual simulation software. Once integrated, the model modules should be accessible through the central simulation architecture to other models and be composable into assemblies of models to represent simulated entities (e.g. a radar, a missile) and complex phenomenon. Models integrated through this architecture should have very limited or no impact on simulation execution speed when compared to native integrated simulation models. The simulation architecture and interface should support tactical messaging in various forms. The proposed simulation architecture should address: • Ease of model integration: A development team with access to specifications of the collaboration interface, but in isolation of the simulation architecture team, should be able to integrate a model into the simulation architecture. Adding a new model should not generally require any changes to the simulation architecture software, with all modifications of data structure (e.g. adding additional parameters) accomplished through model meta-data/parameters. Models needed to produce required additional data items of a new model would themselves be added through the collaboration interface. • Model composability: A team developing a representation of a system or sub-system within the simulation architecture should be able to select needed models from a library of integrated models and build up their respective relationships relative to the representation, to include triggering mechanisms, and association with self and other entities. • Speed of execution: Ideally, a simulation constructed with this architecture would execute as fast as a fully integrated simulation using the same algorithmic models. The difference between this ideal and the developed capability should be minimized. • Ease of data collection and retrieval: All representation/entity level state data and any data passed between models should be collectable. All data collected should be retrievable through a single point of entry by an analyst.
PHASE I: Develop the proposed simulation architecture/collaboration layer to a sufficient level to provide a proof-of-concept. Evaluate its potential value against stated goals, either analytically, through test, or both. Define the software architecture for further development in Phase II.
PHASE II: Develop the proposed software architecture to the level of an operational prototype. Test the prototype for ease of model integration, model composability, speed of execution, data collection, and retrieval. Software must meet government cyber security standards to allow government testing of the prototype. Provide basic documentation of the collaboration interface, algorithms, code structure, coding standards, cyber security scans and practices, and user guidance. Define the expected Phase III development effort required to transition from a prototype capability to an operational capability.
PHASE III: Extend development of the prototype into a fully operational simulation architecture environment with supporting tools as required by concept. All software must meet government cyber security standards for government operational use. Develop documentation of the collaboration interface, algorithms, code structure, coding standards, cyber security scans and practices, and user guidance.
1: Gianni D, D’Ambrogio A, Iazeolla G. A Layered Architecture for the Model-Driven Development of Distributed Simulators, Institute for Computer Science, Social-Informatics and Telecommunications Engineering, March 200 https://dl.acm.org/citation.cfm?id=141629
2: Steinman J, Walter B, Park J, Delane N. A Proposed Open System Architecture for Modeling and Simulation, Joint Program Executive Office for Chemical and Biological Defense, March 200 https://www.sisostds.org/DesktopModules/Bring2mind/DMX/Download.aspx?Command=Core_Download&EntryId=27797&PortalId=0&TabId=10
3: Steinman J, Walter B, Park J, Delane N. A Proposed Open System Architecture for Modeling and Simulation, Joint Program Executive Office for Chemical and Biological Defense, March 200 https://www.sisostds.org/DesktopModules/Bring2mind/DMX/Download.aspx?Command=Core_Download&EntryId=27797&PortalId=0&TabId=10
KEYWORDS: Model, Simulation, Interface, Software, Integration, Federation, Composability, Distributed Objects, Simulation Architecture, Simulation Interfaces, Collaboration Layer