41-50 of 61 results
-
Reconfigurable Guidance and Control Systems for Emerging On-Orbit Servicing, Assembly and Manufacturing (OSAM) Space Vehicles
PI Morad Nazari
CO-I Kadriye Merve Dogan
In this project, the ControlX team with ERAU partnership will develop agile, reconfigurable, and resilient dynamics and G&C algorithms for on-orbit servicing to capture a broad set of OSAM applications such as remediation of resident space object (RSO) (e.g., via de-orbiting, recycle, end-of-life servicing, satellite refueling, etc.) using effective tools and methods involving geometric mechanics, constrained G&C synthesis, and reconfigurable robotic manipulators (RRMs).
Dynamic response to emergent situations is a necessity in the on-orbit servicing, assembly and manufacturing (OSAM) field because traditional on-orbit guidance and control (G&C) cannot respond efficiently and effectively to such dynamic situations (i.e., they are based on either constant mass or diagonal matrix of inertia). In these circumstances, the current challenge is to develop modeling strategies and control systems that exhibit intelligence, robustness and adaptation to the environment changes and disturbances (e.g., uncertainties, constraints and flexible dynamics). Note that the current state-of-the-art methods do not offer a reliable, accurate framework for real-time, optimal accommodation of constraints in the system dynamics that account for orbital-attitude coupling in the motion of the bodies without encountering singularity or non-uniqueness issues. In this project, the ControlX team with ERAU partnership will develop agile, reconfigurable, and resilient dynamics and G&C algorithms for on-orbit servicing to capture a broad set of OSAM applications such as remediation of resident space object (RSO) (e.g., via de-orbiting, recycle, end-of-life servicing, satellite refueling, etc.) using effective tools and methods involving geometric mechanics, constrained G&C synthesis, and reconfigurable robotic manipulators (RRMs).
The proposed work in this Phase I includes reconfigurable systems for on-orbit servicing, assembly and manufacturing with learning control methods that minimize tracking error of the end-effector of the RRM in the presence of uncertainties, optimize configuration and accommodate constraint-changing scenarios. Our developments will avoid singularity; not rely on the concept of costates or Lagrange multipliers that are restrictive; handle system uncertainties while enforcing the constraints; use RRMs in different tasks (recycle, debris removing, maintenance, etc.); not need in sensors or exact model knowledge for robotic arms. Specifically, constrained space vehicle control (predicated on Udwadia and Kalaba (UK) formalism) will offer not only accurate and resilient design but also reconfigurability in that G&C algorithms can easily be modified to suit a wide spectrum of OSAM applications. ControlX team will also consider the feasibility of hardware implementation. The selection of sensors, actuators and onboard computers will be an important trade-off between among size, weight, and power (SWaP) constraints, reliability, and computing performance. A real-time operating system (RTOS) is planned to meet timing and memory management constraints, partitioning hardware resources to control software application interactions. An analysis suite for the autonomous system implementation, based on system modeling and learning techniques, will provide onboard analysis, enabling decision-making as to whether the system can continue to meet mission requirements.
Categories: Faculty-Staff
-
Investigate Detect and Avoid Track Classification and Filtering
PI Richard Prazenica
CO-I Troy Henderson
CO-I Morad Nazari
CO-I Tyler Spence
This research will identify key sources of uncertainty in representative detect and avoid architectures and assess the downstream risks and effects of spurious information on downstream system performance
In this project, which is funded by the FAA ASSURE program, the research team consisting of The Ohio State University, Embry‑Riddle Aeronautical University, Mississippi State University, University of North Dakota and Cal Analytics will work together to:
- Identify the key sources of misleading surveillance information produced by airborne and ground-based detect and avoid (DAA) systems. Develop risk modeling and analysis tools to assess the system-wide effects of false or misleading information on alerting and separation, as well as impacts on pilots in command (PIC) and air traffic operators.
- Provide guidance and recommendations for track classification and filter performance and safety requirements to standards bodies, including Radio Technical Commission for Aeronautics (RTCA) and American Society for Testing and Materials (ASTM) DAA working groups, and inform Federal Aviation Administration (FAA) rulemaking on DAA operations.
Current guidance provided by the Federal Aviation Administration has made beyond visual line of sight (BVLOS) missions an executive priority. Key to the success of these missions is the development of DAA systems capable of providing accurate pilot in the loop, or autonomous deconfliction guidance. Current standards for DAA services provided by RTCA and ASTM do not address the requirements for system performance with respect to generation of false or misleading information to the PIC or autonomous response services of the unmanned aircraft system. This research will identify key sources of uncertainty in representative DAA architectures and assess the downstream risks and effects of spurious information on downstream system performance. Additionally, recommendations will be developed for track classification accuracy requirements that provide sufficient safety margins for enabling DAA services in support of BVLOS missions.
Categories: Faculty-Staff
-
Researching How You Teach Holistic Modeling (RHYTHM)
PI Kelsey Rodgers
CO-I Matthew Verleger
CO-I Lisa Davids
"Models are a critical part of the analysis and design of engineered systems. The purpose of multiple types of models (physical, mathematical, computational, and financial) is to provide a simplified representation of reality that mimics the features of the engineered system, and that predicts the behavior of the system. This project, a collaboration between Embry-Riddle Aeronautical University, San Jose State University, and the University of Louisville, aims to improve engineering students' modeling competence. The project plans to achieve this goal by transforming first-year engineering courses to teach modeling as an engineering tool. The project will change existing course materials, pedagogy, and assessment methods across the three institutions. Each institution will implement its own specific strategy to teach mathematical, physical, computational, and financial modeling, thus providing three different approaches. By comparing student's modeling abilities across the institutions and approaches, the project aims to identify the most impactful approaches for teaching multiple modeling in introductory undergraduate engineering courses.
The project is guided by a "holistic modeling perspective" theoretical framework, that builds on the successful "Models and Modeling Perspective" and "Computational Adaptive Expertise" frameworks. The objectives of the project are to: (1) implement, test, and refine holistic modeling environments for institutions that have flexibility in changing curriculum and for instructors that have different degrees of interest in changing their course(s); (2) implement, test, and refine methods to assess students' modeling abilities; and (3) evaluate and present the results of modeling abilities attained by students at three different universities. A unified language and discussion around modeling will be adopted in all revised courses. An assessment tool to measure students' modeling competence will be developed and implemented at each university. This work builds upon existing research in the development of more easily adaptable and adoptable modeling pedagogies and modeling languages. The following broad research question guides the research: How do students' definitional knowledge, ability to apply, and ability to create models change based on different degrees of modeling integration in the classroom?
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria."
Categories: Faculty-Staff
-
Machine Learning for Dynamic Airspace Configuration towards Optimized Mobility in Emergency Situations
PI Houbing Song
Categories: Faculty-Staff
-
NSF REU Site: Swarms of Unmanned Aircraft Systems in the Age of AI/Machine Learning
PI Houbing Song
CO-I Richard Stansbury
Embry-Riddle Aeronautical University establishes a new Research Experiences for Undergraduates (REU) Site to engage participants in research in drone swarms. The emerging concept of drone swarms, which is defined as the ability of drones to autonomously make decisions based on shared information, creates new opportunities with major societal implications. However, future drone swarm applications and services pose new networking challenges. A resurgence of Artificial Intelligence and machine learning research presents a tremendous opportunity for addressing these networking challenges. There is an overwhelming need to foster a robust workforce with competencies to enable future drone swarm applications and services in the age of AI/machine learning.
The project establishes a new Research Experiences for Undergraduates (REU) Site with a focus on networking research for drone swarms in the age of AI/machine learning at Embry-Riddle Aeronautical University. The goals of the REU Site are: (1) attract undergraduate students to state-of-the-art drone swarm research, especially those from underrepresented groups, and from institutions with limited opportunities; (2) develop the research capacity of participants by guiding them to perform research on drone swarms; (3) grow the participants’ technical skills to enable a wide variety of beneficial applications of drone swarms; (4) promote the participants’ integrated AI/machine learning and drone swarm competencies; and (5) prepare participants with professional skills for careers. The focus of the REU Site is on the design, analysis and evaluation of innovative computing and networking technologies for future drone swarm applications and services. To be specific, research activities will be conducted in three focus areas, notably dynamic network management, network protocol design, and operationalizing AI/machine learning for drone swarms. Each year eight undergraduate students will participate in a ten-week summer REU program to perform networking research for drone swarms under the guidance of research mentors with rich experiences in AI/machine learning and drone swarms. This REU site is expected to foster workforce knowledge and skills about developing new computing and networking technologies for future drone swarm applications and services. This site is supported by the Department of Defense ASSURE program in partnership with the NSF REU program.
Categories: Faculty-Staff
-
A Curriculum Wide Software Development Case Study
PI Massood Towhidnejad
CO-I Thomas Hilburn
This NSF funded research develops case studies of software development for use in software engineering and computing instruction.
Products include realistic projects, complete artifacts throughout the software development life cycle, case studies decoupled from a particular textbook, and case modules designed with varying complexity allowing for use in multiple classes throughout undergraduate and graduate curricula.Categories: Faculty-Staff
-
Encouraging Students to Pursue an Engineering Education and Career
PI Massood Towhidnejad
This NSF-sponsored project provides scholarship for engineering students pursuing degrees in computer science, computer engineering, electrical engineering, mechanical engineering and software engineering.
Working closely with faculty and student mentors, scholarship recipients are involved in multi-disciplinary projects involving unmanned and autonomous systems throughout their four years of undergraduate study.Categories: Faculty-Staff
41-50 of 61 results