Scalable Cyber-Physical Simulation for Automated Cyber Agent Training

Modern cyber-physical systems (CPS) are highly complex systems-of-systems, in which understanding the breadth and severity of cyberattacks is highly challenging. As cyberattacks and defensive operations become increasingly automated, there is a greater need to understand the complexities of interactions between the cyber and physical worlds. A scalable, detailed simulation platform will provide a means of developing and evaluating automated techniques within these complex systems. We will leverage the Cyber-Physical Systems Wind Tunnel (CPSWT), a model-based framework for rapidly synthesizing heterogeneous simulation integration, to develop a standard method for assessing cyber operations within these systems. We will leverage the Cyber AI Gym Environment (CAGE), an APL developed tool, to assess the feasibility of Reinforcement Learning (RL) within the integrative simulation environment, demonstrating the ability to select and target attacks based on the secondary observables generated in non-cyber simulations. The overarching project goal is to increase adaptability of automated learning solutions by:

  • Showing that we can learn offensive and defensive maneuvers simultaneously within realistic simulation environments,
  • Showing that we can customize, deploy, and validate general RL policies to highly realistic modeling and simulation (and to the real-world using hardware-in-the-loop simulation), and
  • Showing that high-fidelity simulation provides a viable approach towards better understanding and validating cyber effects.

Key Outcomes

This project is the first award as part of the new relationship that Himanshu developed with APL that enables Vanderbilt PIs to submit proposals collaboratively with APL PIs to APL's internal research and development (IRAD) missions on a yearly basis. Once funded, the project usually will gain options increasing budget if necessary and for continuation beyond the initial year.

Sponsor: John Hopkins University - Applied Physics Laboratory (JHU-APL)
Lead PI: Himanshu Neema

Benching Computer Vision Algorithms for Basketball

Sponsor: Noah Basketball
Lead PI: Jules White

Science Projects Integrating Computing and Engineering (SPICE)

SPICE curriculum materials blend disciplinary core ideas, science and engineering design practices, and crosscutting concepts as called for in the the Next Generation Science Standards (NGSS). We incorporate principles of evidence-centered design, knowledge integration, and informed engineering design to develop project-based curriculum materials that promote the integration of science, engineering, and computational thinking (CT).

Sponsor: NSF
Lead PI: Gautam Biswas

Multi-level Learner Modeling for Land Navigation Training Applications

Sponsor: ARL
Lead PI: Gautam Biswas

Rapid Scenario-Driven Integrated Simulation Experimentation Framework

Cyber-Physical Systems (CPS) are composed of a wide range of networked physical, computational, and human/organization components. These systems are highly complex as they have many different heterogeneous components, such as physical, computational, and human. Simulation-based evaluation of the behavior of CPS is complex, as it involves multiple, heterogeneous, interacting domains. Each simulation domain has sophisticated tools, but their integration into a coherent framework is a difficult, time-consuming, labor-intensive, and error-prone task. This means that it is difficult to conduct computational studies rapidly and provide timely answers to the planners, operators, and policy makers. Furthermore, CPS behavior has to be tested against a number of scenarios and situations, meaning that a large number of simulations must be executed covering the entire space of possibilities. This project leverages our Cyber-Physical Systems Wind Tunnel (CPSWT) framework -- that enables rapid, model-based integration of a variety of simulation tools -- to develop methods, tools, and approaches for creating a scenario-driven experimentation environment that can support rapid investigation of CPS using a large combination and variants of experiment scenarios. The objectives of this project are to develop: (1) Scenario-driven experimentation capability of the CPS simulation integration framework by supporting the modeling, parameterization, configuration, execution, and monitoring of integrated simulation experiments; and (2) Capability to enable instrumentation of experiment data-generation dynamically (i.e., at run-time) according to the requirements of scenarios designed to experiment with integrated simulations -- which will enable generation of pertinent datasets for conducting specific analyses as well as training Artificial Intelligence (AI) algorithms for detecting and analyzing rare events in the simulations.

Key Outcomes

The project has been successfully transitioned to the Communications Technology Laboratory (CTL) within the US National Institute of Standards and Technology (NIST).

Sponsor: National Institute of Standards and Technology
Lead PI: Himanshu Neema

Implementing Betty's Brain in the PILA Environment

Sponsor: OECD
Lead PI: Gautam Biswas
Co-PI: Marian Rushdy

Cyber Makerspace - Science of Security for Cyber-Physical Systems Lablet

Makerspaces are very popular because they provide a hands-on experience for young learners to experiment with technology. One drawback is that the focus of educational experiences in makerspaces are necessarily on the hardware. Computing aspects, especially more advanced concepts such as cybersecurity, take a back seat. We will team up with Martin Luther King Jr. Academic Magnet School (a public school with 60% minority student population in Nashville) to pilot a cyber makerspace, where students build virtual robots, including advanced sensors and actuators that they would never have access to in a physical makerspace, and the virtual worlds the robots live in. In addition, students will need to implement the desired behavior of the robots to solve challenges including ones related to cybersecurity. The cyber makerspace will make it possible to teach advanced concepts in a practical and hands-on, yet playful manner that will result in high level of engagement and consequently, highly effective learning.

This project builds upon our prior work with NetsBlox, an open source, browser-based visual programming environment and corresponding cloud-infrastructure. NetsBlox integrates distributed programming capabilities at a level accessible for novice programmers through two conceptually simple, yet powerful abstractions: Remote Procedure Calls (RPCs) and message passing. They enable students to create engaging projects such as programs that access online data source and services such as Google Maps, weather data, stock quotes and many more, as well as distributed programs such online multiplayer games or a chatroom. 

Networked physical devices can also be accessed using the same abstractions. For example, students get to write programs to remotely control WiFi-enabled robot vehicles in a setting where other students can intercept the wireless commands and hijack each other’s robots. This practice provides the motivation and a physical testbed to teach cybersecurity in a hands-on, practical manner. NetsBlox also enables Google Docs-like collaboration. In this shared synchronous online environment, students can work on a common project from their own computers regardless of their geographic location. This type of collaboration results in rich and diverse opportunities that have been shown to improve the perceptions, confidence and performance of students underrepresented in STEM.

Sponsor: Department of Defense
Lead PI: Akos Ledeczi

Collaborative Research: Beyond CS Principles: Engaging Female High School Students in New Frontiers of Computing

Building on the foundations set by the AP Computer Science (CS) Principles course, this project seeks to dramatically expand access, especially for high school girls, to the most exciting and emerging frontiers of computing, such as distributed computation, the internet of things (IoT), cybersecurity, and machine learning, as well as other 21st century skills required to productively leverage computational methods and tools in virtually every profession. Creating pathways that stimulate high school learners' interest in advanced topics with the goal of building a diverse, gender-balanced, future-ready workforce is a crucial and impactful imperative addressed in this work. An experienced multi-disciplinary team of researchers, high school teachers, and industry partners will be involved in the design of a new, modular, open-access curriculum called Computer Science Frontiers (CSF) that provides an engaging introduction to these advanced topics in high school (that are currently accessible only to CS majors in college). To address the dire gender disparity in computing, the project will design and research the curriculum to engage female students. Through studying the impact of innovative computing tools and curricular units on learning, attitudes, interests, and collaboration of students (and especially young women), the project will advance discovery and understanding to aid the cause of broadening participation in technology-related careers as well as the future of work at the human-technology frontier. The open access modular design of the curricular materials and the dissemination activities will ensure wide adoption of the products of this research.

The project leverages NetsBlox, a powerful yet easy-to-use visual programming environment that has been shown to increase engagement and interest in computing. Additionally, NetsBlox supports effective collaboration while facilitating learning of distributed computing, networking, and cybersecurity in informal settings. Early phases of the project will involve design and refinement of curricular modules through co-design workshops with seven participating teachers, summer camps with students, and pilot implementations in Tennessee and North Carolina, leading to full classroom implementations in Year 3 by the high school teachers. Data collected through surveys, student portfolios, teacher interviews, and digital logs from NetsBlox will be analyzed. Research findings will help in understanding if and how advanced computing methods alongside key competencies can be introduced in high school; how pedagogies involving project-based activities around real-world, multidisciplinary problems work to increase female students' interests in computing; and which advanced topics work better than others in terms of difficulty level and engagement.

This project is funded by the Innovative Technology Experiences for Students and Teachers (ITEST) program, which supports projects that build understandings of practices, program elements, contexts and processes contributing to increasing students' knowledge and interest in science, technology, engineering, and mathematics (STEM) and information and communication technology (ICT) careers.

Award Number: 1949472
Sponsor: National Science Foundation
Lead PI: Akos Ledeczi
Co-PI: Brian Broll

Collaborative Research: Computational Modeling for Integrating Science and Engineering Design (CMISE): Model Construction, Manipulation, and Exploration

Computational Modeling for Integrating Science and Engineering Design (CMISE) will conduct a series of experiments to systematically compare different computational modeling activities on 5th and 6th grade students’ engineering design processes, their understanding of engineering, science and computational thinking concepts, as well as science teachers’ confidence and ability to implement integrated STEM and computing curricula. Computational modeling involves a high cognitive load, and research to date is unclear whether the payoff primarily entails learning computing or whether students’ science and engineering learning benefit as well. CMISE will investigate how different types of computational modeling activities promote integrated student learning of science and engineering. CMISE will have immediate impacts on STEM + Computing offerings for the Metro Nashville Public School district where the project will be conducted; broadly it will also help strengthen and grow a diverse STEM workforce by bringing authentic and compelling science and engineering opportunities to fifth and sixth grade students. This project will also provide designers and researchers with empirical evidence for how to effectively integrate computer modeling with science and engineering learning activities in different settings. The CMISE curriculum and teacher support materials will be made freely available through project website, allowing these resources to reach a wide range of teachers beyond those included in the study.

CMISE will leverage a previously developed and refined Next Generation Science Standards-aligned curriculum unit that integrates the Earth Science concept of urban water runoff with a meaningful engineering design problem for fifth- and sixth grade students to conduct fundamental research to better understand how different types of computational modeling activities mediate the connections between science and engineering learning. CMISE will conduct a series of design experiments to systematically compare the affordances of three computational modeling activities on students’ engineering design processes, their understanding of engineering, science and computational thinking (CT) concepts and practices, as well as science teachers’ confidence and ability to implement integrated STEM and computing curricula. The three activities being compared are computational model (CM) construction (where students model a science phenomenon in a given programming language), CM manipulation (where students inspect either the code or the simulation for a given CM of a science phenomenon), and CM exploration (where students explore a given simulation of a science phenomenon without viewing the underlying code). CMISE adopts a strong theoretical framing and a systematic design experiment approach to contribute to the learning theory of how students interact with and learn using different types of computational modeling activities. It will apply quantitative and qualitative analysis methods to combine established statistical analysis methods with novel analytics approaches and derive relations between students' learning behaviors and performance in the three experimental conditions. This design experiment studies will disentangle how these conditions influence the synergistic learning of science, engineering, and CT.

This project is co-funded by the EHR Core Research (ECR) and CS for All: Research and RPPs programs. ECR supports work that advances fundamental research on STEM learning and learning environments, broadening participation in STEM, and STEM workforce development.

Award Number: 2055609
Sponsor: NSF
Lead PI: Gautam Biswas

Collaborative Research: An Interdisciplinary Approach to Prepare Undergraduates for Data Science Using Real-World High Frequency Data

With support from the NSF Improving Undergraduate STEM Education Program: Education and Human Resources (IUSE: EHR), this project aims to serve the national interest by improving undergraduate understanding of data science. It will accomplish this goal by incorporating data science concepts and skill development in undergraduate courses in biology, computer science, engineering, and environmental science. Through a collaboration between Virginia Tech, Vanderbilt University, and North Carolina Agricultural and Technical State University, the project will develop interdisciplinary learning modules based on high frequency, real-time data from water and traffic monitoring systems. The project intends to develop a common approach for introducing data science concepts in STEM disciplinary courses. By embedding data science into a variety of undergraduate STEM courses and creating a partnership that includes a Historically Black College/University, this project has the potential to broaden participation in data science, including participation of students from populations that are underrepresented in data science and/or STEM fields.

This project will develop data science learning modules to implement in eight existing STEM courses at the collaborating institutions. The learning modules will be motivated by real-world problems and high-frequency datasets, including a water monitoring dataset from Virginia Tech, and transportation and building monitoring datasets from Vanderbilt. The learning module topics will include: Interdisciplinary Learning, Data Analytics, and Industry Partnerships. These topics will facilitate incorporation of real-world data sets to enhance the student learning experience and they are broad enough that they can incorporate other data sets in the future. The project aims to develop and implement an interdisciplinary collaborative approach to support undergraduate students in developing data science expertise through their disciplinary course work. Such expertise will better prepare students to enter the STEM workforce, especially those STEM professions that focus on smart and connected computing. The project will investigate how and in what ways the modules support student learning of data science. The project will also investigate how implementation of the modules varies across the collaborating institutions. It is expected that the project will define key considerations for integrating data science concepts into STEM courses and will host workshops to introduce faculty to these considerations and strategies so they can incorporate the learning modules into the STEM courses that they teach. The project collaborators will provide the framework for generalizing and transferring the learning modules to other STEM education communities, thus broadening the scope and the impact of this project beyond the three collaborating institutions. The NSF IUSE: EHR Program supports research and development projects to improve the effectiveness of STEM education for all students. Through the Engaged Student Learning track, the program supports the creation, exploration, and implementation of promising practices and tools.

Award Number: 1915538
Sponsor: NSF
Lead PI:
Co-PI: Abhishek Dubey
Subscribe to Learning and Training Environments