DDDAS 2020 Conference
The research advances and S&T capabilities showcased at DDDAS2020, stem from the DDDAS (Dynamic Data Driven Applications Systems) paradigm, whereby instrumentation data are dynamically integrated into an executing application model while in reverse,the executing model controls the instrumentation. DDDAS plays a key role in creating capabilities in many application areas and is driving advances in foundational methods, through system-level (as well as subsystems-level) representation, that includes comprehensive principle-and physics-based-models and instrumentation, uncertainty quantification, estimation, observation, sampling, planning and control.
This website is dedicated to showcasing scientific and technological advances in complex systems modeling and instrumentation methods enabled under the rubric of the Dynamic Data-Driven Applications System paradigm. The website is a scientific community forum with hyperlinks to DDDAS research projects, virtual proceedings, related software, announcements and other news.
As articulated by Dr. Darema who pioneered the DDDAS paradigm: “in DDDAS instrumentation data and executing application models of these systems become a dynamic feedback control loop, whereby measurement data are dynamically incorporated into an executing model of the system in order to improve the accuracy of the model (or simulation), or to speed-up the simulation, and in reverse the executing application model controls the instrumentation process to guide the measurement process. DDDAS presents opportunity to create new capabilities through more accurate understanding, analysis, and prediction of the behavior of complex systems, be they natural, engineered, or societal, and to create decision support methods which can have the accuracy of full-scale simulations, as well as to create more efficient and effective instrumentation methods, such as intelligent management of Big Data, and dynamic and adaptive management of networked collections of heterogeneous sensors and controllers. DDDAS is a unifying paradigm, unifying the computational and instrumentation aspects of an application system, extends the notion of Big Computing to span from the high-end to the real-time data acquisition and control, and it’s a key methodology in managing and intelligently exploiting Big Data”
The DDDAS paradigm, and opportunities and challenges in exploiting the DDDAS paradigm have been discussed in a series of workshops, starting in 2000. The reports from these workshops, included in the present website, identified new science and technology capabilities, driven by and enabled through the DDDAS paradigm, and these include new modeling methods, algorithms, systems software, and instrumentation methods, and well as the need for synergistic multidisciplinary research among application domains researchers, and researchers in mathematics, statistics, and computer sciences, as well as well as researchers involved in design and development of instrumentation systems and methods. In addition to these workshops, through corresponding government sponsored initiatives, research efforts commenced to address the challenges and create new capabilities. As shown through the increasing body of work, DDDAS is applicable to many areas, such as civil engineering, aerospace, manufacturing, transportation, energy, medical diagnosis and treatment, environmental, weather, and climate, etc. This website presents examples of advances through DDDAS in areas and aspects identified above. Additional information
DDDAS Advantages From High-Dimensional Simulation
DDDAS Challenges and Processes
The key developments of the integration of the instrumentation, models, and software to enable the development of DDDAS include theory, algorithms, and computation. The theory includes mathematical advances (retrospective cost modeling – check); while the algorithms support new paradigms (e.g., ensemble Kalman filter, Particle filter, optimization techniques). The computation methods align with the developments in the continuing networked society such as non-convex optimization and data flow architectures.
The challenges DDDAS seeks to advance include data modeling, context processing, and content application. To bring together data, context and content requires addressing issues in model fidelity such as how many parameters are needed for system control. When data is collected, it needs to be preprocessed to determine whether its inherent information matches the context. One example includes clutter reduction, sensor registration, and confuser analysis in vehicle tracking. Finally, another key challenge is that of sampling. Sampling is the multiresolution needed to monitor the situation, environment and network context to explain the content desired.
DDDAS Book I
The Handbook of Dynamic Data Driven Applications Systems establishes an authoritative reference of DDDAS, pioneered by Dr. Darema and the co-authors for researchers and practitioners developing DDDAS technologies.
Call for Chapter Submissions for DDDAS-Book-II — Handbook on Dynamic Data Drive Application Systems (DDDAS) (Vol. II)
DDDAS In The News
- Researchers at RPI study causes of fatal air crashes: Team working on system to recognize when sensors fail.
Dynamic Data-Driven Decisions: Application to a Self-Aware UAV. We are developing the methods and algorithms that enable (1) on-board dynamic decision-making for an autonomous aerial vehicle, (2) creation of a vehicle Predictive Digital Twin, and (3) illustration of the benefits of Digital Thread. The distinguishing aspect of our approach is that it is both data-driven and physics-based. Our hypothesis is that by leveraging both physics-based knowledge (through physics-based simulation data and physics-based reduced models) and dynamic data (through on-board sensors) we can issue better decisions than if we were to use data alone.
- Developing a digital twin [ details ]
- The PILOTS Programming Language: PILOTS is a Programming Language for spatiO-Temporal data Streaming applications, especially designed to be used for building applications that run on moving objects such as airplanes, cars, and so on. With very high-level specifications, users can easily build applications that take spatio-temporal data streams as an input and produce streams as outputs for use by other applications such as actuator controls, data mining/analyses/visualization, and error correction codes. These applications can be treated as components of a larger stream processing system.
Grant Awarded to Increase Intelligence in Aerospace Systems: “We believe the next generation of aerospace systems will be able to feel, think, and react in a way that is similar to how biological systems behave.”
RIT researchers developing ways to use hyperspectral data for vehicle and pedestrian tracking. A classic scenario plays out in action films ranging from Baby Driver to The Italian Job: criminals evade aerial pursuit from the authorities by seamlessly blending in with other vehicles and their surroundings. The Air Force Office of Scientific Research (AFOSR) has Rochester Institute of Technology researchers utilizing hyperspectral video imaging systems that make sure it does not happen in real life.
At SC19: Developing a Digital Twin: In the not too distant future, we can expect to see our skies filled with unmanned aerial vehicles (UAVs) delivering packages, maybe even people, from location to location.