NEWSLETTER Quarter 4, 2018
Testing Evaluation and Control of Heterogeneous Large-scale Systems of Autonomous Vehicles
(TECHLAV)
contents Open letter from the Director ................................................................................................. 3
Message from the Dean .................... ....................................................................................... 4
About TECHLAV ................................ ....................................................................................... 5
Research Thrust highlights ...................................................................................................... 6
Message from the SIAB Chair ................................................................................................... 9
Education and Outreach & Publication Highlights .................................................................. 10
TECHLAV Team ....................................................................................................................... 12
SIAB Members ....................................................................................................................... 14
Student spotlight .................................................................................................................. 15
Upcoming Events .................................................................................................................. 16
Acknowledgement ................................................................................................................ 17
TECHLAV Newsletter is a quarterly publication of the Autonomous Control and Information
Technology institute, North Carolina Agricultural & Technical State University.
Open letter from the Director
TECHLAV Center News Page 3
The Testing, Evaluation and Control of
Heterogeneous Large-scale Systems of
Autonomous Vehicles (TECHLAV) Center’s
vision is to serve as a recognized national
leader in control and autonomy and to develop
tools, techniques, and algorithms for the
modeling, analysis, control, coordination
testing, and evaluation of large-scale systems
of autonomous vehicles.
The goal of this newsletter is to highlight and
celebrate the many achievements of this team
and as such, each quarter we will spotlight a
TECHLAV faculty and student.
TECHLAV continues to move closer to
producing substantial work that will increase national and regional economic growth fostered by the
development, adoption, and effective use of new technologies. This year has been a reminder that we
have been able to make tremendous strides thanks to the commitment to partnership, dedication and hard
work of our faculty, staff members, and most importantly students at North Carolina A&T State
University, the University of Texas at San Antonio, and the Southwestern Indian Polytechnic Institute.
The support of our sponsors, scientific and industrial advisory board members, administrators and
supporting teams at all three institutions have also ensured our success. Through the efforts of many that
we have made progress toward longstanding advances that we all desire. Thank you for your
commitment to TECHLAV’s vision and success!
Abdollah Homaifar, Ph.D.
NASA Langley Distinguished Chair Professor,
Duke Energy Eminent Professor and
Director, Autonomous Control and Information
Technology (ACIT) Institute,
Director, TECHLAV Center
Email: [email protected]
Telephone: 336-285-3271
message from the Dean If you have not had a chance to visit the heterogeneous autonomous vehicle testing facility, which is one of the core labs of the TECHLAV team, I encourage you to do so. This high-bay area laboratory enables both aerial and ground-based autonomous vehicles to be simultaneously and safely operated; and it continues to evolve to effectively house the work the TECHLAV team is advancing. I applaud the achievements and agility of the TECHLAV team, and I sincerely thank the center’s sponsors, and the center’s scientific and industrial advisory board for your continued support and dedication. Because of all of you, TECHLAV’s impact in creating groundbreaking technical advancements and in providing a well-prepared workforce in the area of autonomous systems, will continue to expand. TECHLAV and the center’s achievements are possible because of the engagement and collaborations of its team members and supporters. I congratulate all who are contributing to its success. For those excited by the future opportunities for this exciting center, I encourage you to engage with the center and keep abreast of its’ many groundbreaking activities. Congratulations to the TECHLAV Team! Sincerely, Robin N. Coger, Ph.D Dean, College of Engineering North Carolina A&T State University
Congratulations to the faculty, staff, and students of the Testing, Evaluation and Control of Heterogeneous Large-scale Systems of Autonomous Vehicles (TECHLAV) Center, led by Dr. Abdollah Homaifar, for all the center is accomplishing in the important area of autonomous systems. TECHLAV relies on a multidisciplinary team of experts from multiple institutions. The center’s inclusion of diverse thinkers in creating solutions in control and autonomy has made it an excellent demonstration of the strength and possibilities of collaboration. TECHLAV is also directly aligned with our College of Engineering’s focus on cultivating leaders who excel in academics, innovation, and research. The center has prioritized the production of quality scholarly output. In 2018 alone, its researchers published over 23 relevant conference and journal papers and filed two patents. Additionally, because TECHLAV is a strong supporter of students, its research and educational activities continue to create a fertile ground for developing a well-prepared future autonomous system workforce. Currently, the center supports 24 doctoral students, 8 master’s degree students, and 15 undergraduate students. I am confident that these engineers and computer scientists will contribute to our society in ways we can only imagine.
TECHLAV Center News Page 4
about TECHLAV
Our Mission The TECHLAV Center is (1) addressing fundamental problems in m o d e l i n g , a n a l y s i s , c o n t r o l , coordination, test and evaluation of autonomous systems (2) serves as a national resource in education and research in Autonomy, (3) provides outreach services in autonomy related areas, and (4) fosters linkages among national institutions of higher education, government agencies and private companies (5) commercializes TECHLAV technologies for the benefit of the national economy.
Our Vision The TECHLAV Center will serve as a recognized national leader in control and autonomy to develop tools, techniques, and algorithms for m o d e l l i n g , a n a l ys i s , co n t ro l , coordination, testing, evaluation of large scale systems of autonomous vehicles, and to provide advanced education in autonomy related areas.
The TECHALV Center is a Center of Excellence in Autonomy, funded by the Department of Defense. The Center, led by N.C. A&T is comprised of a strong multi-disciplinary team and synergizes expertise in Control, Communication, and Human Cognition from N.C. A&T and its collaborator, the University of Texas San Antonio (UTSA). The Center also partners with the Southwestern Indian Polytechnic Institute (SIPI) to provide and promote education, outreach activities, and curriculum development to a larger Native American community.
The research activities are addressing two fundamental grand challenges: 1) Teaming and Cooperative Control of Large-scale Autonomous Systems of Vehicles (LSASVs) integrated with human operators, and 2) Testing, Evaluation, Validation, and Verification of LSASVs.
These challenges are met through three interleaved research thrusts: • Thrust 1: Modeling, Analysis and Control of LSASVs • Thrust 2: Resilient Control & Communication of LSASVs
• Thrust 3: Testing, Evaluations and Verification of Large-scale Autonomous Vehicles (TEVLAV)
TECHLAV Center News Page 5
Photo credit: www.eduardosuastegui.com
Demonstration, Implementation and Integration (DII) - Test and validates the Performance of the developed tools and protocols in real world scenarios .
Designed a hybrid controller for the real-
time control and path planning of an
attacking autonomous vehicle to win a
dynamic reach-avoid game against a
defending vehicle.
Hardware experiments testing two
configurations of the cloud-edge
infrastructure have been performed.
These experiments were used to verify
that the cloud and edge device
configurations are working correctly for
proper data collection.
Statistics gathering using hardware
configurations will be performed in the
next quarter on the cruise ship disaster
scenario response simulator. Software
modifications have to been performed to
convert the complex cruise ship disaster
scenario simulator to be compatible with
the cloud and edge device architecture.
thrust 1 Highlights
TECHLAV Center News Page 6
Thrust 1 led by Dr. Mo Jamshidi at UTSA, develops scalable methodologies to improve modeling, analysis, localization, navigation, and control of Large-scale Autonomous Systems of Vehicles (LSASVs). The highlights of this quarter are listed as follows:
Developed a quadcopter with capabilities of safe indoor autonomous flight, and automated landing and charging procedures. This quadcopter acts as a complex model for which a basis for making much needed localization improvements to the terrestrial testbed of robotic platforms is provided.
Identified an off-road outdoor capable chassis to
replace the small indoor-only chassis for the
UGV platform. This platform is used for
multiple vehicle research and can be obtained
for less than $200. This new Ackerman style
chassis is a capable, scale-sized rock crawler
platform.
TECHLAV Center News Page 7
thrust 2 Highlights Thrust 2, led by Dr. Ali Karimoddini at NCAT, develops systematic techniques, tools, and algorithms to enhance the reliability and efficacy of the control structure and the communication backbone for Large-scale Autonomous Systems of Vehicles (LSASVs) integrated with human operators in dynamic and uncertain environments such as the battlefield. The highlights of this quarter are listed as follows: Developed an algorithmic approach for
constructing a semi-asynchronous diagnoser
that detects faults in a system despite
uncertainties in information about the initial
conditions of the system, without requiring the
system under diagnosis to be restarted to re-
synchronize the diagnoser and the system
under diagnosis.
Developed a new approach for collecting and
formatting data so that the fault diagnostic
model can accurately make predictions. Once
data is collected, the fault diagnostic model is
developed, trained, and tested using RNN with
LSTM units.
Given a number of ground IoT users dis-
tributed in a 2-D area, an optimization
problem was formulated to simultaneous-
ly optimize the location and number of
the UAVs in order to provide wireless
coverage for the users.
Investigated the impact of inaccurate
beam alignment on the performance of a
directional mmWave D2D network of
communicating agents and used tools
from stochastic geometry to quantify the
signal-to-interference-plus-noise ratio
(SINR) coverage probability in the pres-
ence of beam misalignment, which can be
applied to evaluate D2D network perfor-
mance.
Reconstructed Advantageous Asynchro-
nous Actor-Critic Algorithm (A3C) in
MATLAB that has thus far showed posi-
tive results. Most of the diagnostic data to
Tensorboard has been ported, which is an
ideal way to visualize Neural Network
performance.
TECHLAV Center News Page 8
thrust 3 Highlights Thrust 3 led by Dr. Younho Seong at NCAT,
provides technologies and tools for testing,
evaluation, validation, and verification of
heterogeneous Large-scale Autonomous Systems
of Vehicles (LSASV). The highlights of this quarter
are listed as follows:
Developed a high level Architectural Analysis
and Design Language (AADL) model of a
heterogeneous multi-agent system.
Employed the Reinforcement Learning (RL)
approach for capturing the human-robot
cooperation. For this purpose, we used an MDP
model of the agent to describe the system
(agent and the human operator). We then used
this model to design a controller for the agent
to achieve its goal in collaboration with a
human operator. Since the agent does not have
any prior knowledge about the system, Q-
learning algorithm was employed to solve the
RL-based human-robot collaboration problem.
Created a simple testbed in which the impact of
automation failure may cause disruption in
human operators' reliance on automation
which consequently reflected on human brain
signals captured.
Conducted two experiments in human
cognition, decision making and neural
responses to different information visualization
to investigate how the information display
could impact human decision making process
and performance under uncertain environment
and the results were summarized.
Developed a dynamic feature selection pro-
cedure to enhance the computational effi-
ciency in the sensor data stream analysis of
LSASVs.
Developed a data driven model for target
classification using the theory of belief
functions. The uncertain information with
respect to various attributes of the un-
known target is encoded by belief functions
which are fused to obtain the final belief
function. Decision regarding the class of
the target is taken based on the pignistic
probability distribution of the final mass
function. For proper management of possi-
ble conflicts that may emanate as a result of
combination of belief functions induced
from different sensor data. We introduced
a combination scheme that accounts for not
only the information contained in the indi-
vidual belief functions but also the reliabil-
ity of the source producing them.
TECHLAV Center News Page 9
message from the SIAB Chair
Dr. Edward Tunstel is an Associate Director of
Robotics in the Autonomous and Intelligent Systems
Department of UTRC focused on human-
collaborative robotics and autonomy for aerospace
and building industries. He joined UTRC in 2017
after 10 years at Johns Hopkins Applied Physics
Laboratory as a senior roboticist and as space
robotics and autonomous control lead engaged in
research for national security and space applications.
Prior to APL he was with NASA JPL for 18 years, as
a senior robotics engineer, group leader of its
Advanced Robotic Controls Group, and flight
systems engineer responsible for autonomous
navigation and associated V&V for NASA Mars
rovers. He earned B.S. and M.S. degrees in
mechanical engineering from Howard University
and the Ph.D. in electrical engineering from the
University of New Mexico. He is an IEEE Fellow,
President of the IEEE SMC Society, and NSBE
Professionals member.
Edward Tunstel, Ph.D., FIEEE
Chair, Scientific and Industry Advisory Board
As Chair of the Scientific and Industry Advisory
Board, I have enjoyed the vantage point, shared
with fellow Board members, from which we can
objectively assess its technical direction and offer
advice toward its increased effectiveness and broad
impact. This is particularly exciting at this time in
technology history when we are collectively
witnessing the evolution, and creating the future of
computer-based autonomous systems. We are doing
a great service focusing on problems of interest to
the Department of Defense and shared by other
sectors of industry. The graduates of TECHLAV
will be highly sought after as a result of their
education and training in cutting edge autonomous
systems synthesis and evaluation for real-world
applications.
From the SIAB perspective, TECHLAV is doing
very well at earning its designation as an Autonomy
Center of Excellence and is performing on par with
other major universities engaged in autonomy
research. This should not be conflated with the
longstanding field of test & evaluation for the range
of engineering systems that preceded systems now
embedded with computational autonomy. The latter
present new and formidable challenges for which
advanced solutions are being fueled by algorithms
and methodologies of autonomous control, machine
learning, cognitive computing, and knowledge-
based or automated reasoning facilities of AI.
Within this new context, TECHLAV students and
faculty are advancing the state of the emerging art
as applicable to large-scale systems of autonomous
vehicles. It is great to see North Carolina A&T State
University, the University of Texas at San Antonio,
and the Southwestern Indian Polytechnic Institute
collectively stepping into and owning this role.
highlight Education and Outreach TECHLAV promotes a diverse, productive, and prepared workforce with more STEM graduates: N.C. A&T’s education and outreach is led by Dr. John C. Kelly, and SIPI’s is led by Dr. Nader Vadiee. The Center will strengthen planned academic programs at N.C. A&T, and will increase multidisciplinary and cross-listed courses across these three campuses. Organized bi-weekly seminar series for faculty and students at N.C. A&T, UTSA and SIPI to present,
share ideas and update one another on the progress of the respective tasks. The details can be found
on the TECHLAV Website at the following URL addresses: http://techlav.ncat.edu/seminars.html".
During this period, research assistantship was provided to 18 PhD and 8 Master students, as well as 8
undergraduate students at N.C. A&TSU, UTSA, and SIPI. These students, in collaboration with their
respective faculty advisors are involved in various research tasks and testbed preparations.
Offered 19 autonomy related courses between the three campuses.
The primary outreach focus of this quarter was on 4th-8th grade students across the state of North
Carolina. A total of 15 FIRST LEGO League activities/events were organized, comprising of approxi-
mately 500 excited teams. The students programmed LEGO Mindstorms EV3 robots to autonomously
solve missions using robot attachments built out of LEGOs. The students took advantage of touch,
infrared, gyro, color, and ultrasonic sensors to control their robots. Each team was judged on project
presentations, which were focused on identifying the challenges humans must overcome to travel in
our solar system for long periods and proposing an innovative solution.
TECHLAV Center News Page 10
highlight Education and Outreach A group of ACIT undergraduate students and PhD mentors, worked on establishing wireless com-
munication between Ground IoT Devices and UAVs. The objective of the project was to develop a
way to minimize the total transmit energy from UAVs to UGVs in wireless communication.
Another group of UG students and PhD mentors at the ACIT lab were involved in a 3D Navigation
by UAV project. Multiple programs were installed onto the Intel Aero compute board, which was
the basis for this project. Such programs included a Robotic Operating System (ROS), and OpenCV.
A PCL library was also installed that will be used for obstacle avoidance in the future.
Undergraduate students at the ACCESS laboratory worked towards the testbed development and
implementation of an autonomous car under the Autodrive Challenge organized by SAE Interna-
tional and GM. The car was redesigned more efficiently by using a more appropriate switch and
moving the system away from rely-
ing on a heavy-duty power inverter.
To support the large data transfer
rates, a low powered, more rugged
routing switch capable of higher
bandwidths was used.
The SIPI Pre-Engineering team and
Dr. Nader Vadiee, visited the UNM
Mechanical Engineering Fluid Dy-
namics Lab and met with UNM pro-
fessors Peter Voribie and Randy Tru-
man.
TECHLAV Center News Page 11
publications Published and submitted 21 journal and conference papers (one published conference paper, one accepted journal paper, seven accepted conference papers, six submitted journal papers, and six submitted conference papers) in relevant journal and conference papers as research outcomes of TECHLAV. The list of publications is available at http://techlav.ncat.edu/publications.html"
TECHLAV Team Dr. Abdollah (Ebbie) Homaifar NASA Langley Distinguished Chair Professor, Duke Energy Eminent Professor at
N.C. A&T State University, Director of the Autonomous Control and Information
Technology (ACIT) Institute, and Testing, Evaluation, and Control of Heterogene-
ous Large-scale Systems of Autonomous Vehicles (TECHLAV) Center.
Dr. Mo Jamshidi leads Thrust 1. He is a Fellow of IEEE, ASME, AAAS, NYAS,
TWA. He is also the Lutcher Brown Endowed Professor at the UTSA. He has been
an advisor to NASA, US Air Force, USDOE and EC/EU.
Dr. Ali Karimoddini is the Deputy Director of TECHLAV, Leads research Thrust
2, and is the Director of the Autonomous Cooperative Control of Emergent Sys-
tem of Systems (ACCESS) laboratory at N.C. A&T. He has received over 12 mil-
lion in research support from federal funding agencies and industrial partners.
Dr. Younho Seong leads research Thrust 3. He is an Associate Professor in Indus-
trial & Systems Engineering at N.C. A&T and has performed research on judg-
ment, decision making, and trust in automation for more than 15 years. His goal is
to improve judgment and decision making performance by transforming the deci-
sion aid to be “more transparent.”
Dr. John C. Kelly leads the education and outreach component of the center. He
has served as the chairperson of the ECE Department at N.C. A&T since 2000. Pri-
or to that, he served as the Associate Dean for Graduate and Research Programs
in the College of Engineering.
Dr. Sun Yi obtained his B.S. in Mechanical and Aerospace Engineering from
Seoul National University in 2004, and his M.S. in Mechanical Engineering in 2006
and Ph.D. in Mechanical Engineering in 2009 from the University of Michigan,
Ann Arbor.
TECHLAV Center News Page 12
TECHLAV Team
TECHLAV Center News Page 13
Mrs. Shar Seyedin is TECHLAV’s Program Manager within N.C. A&T’s ACIT
Institute. Her expertise is in program and project management with a focus on
strategic multi-disciplinary team management and multi-agency project organiza-
tion, implementation, and execution. She is a business professional with over 12
years of experience in both financial & academic research sectors.
Dr. Joseph Stephens earned his Ph.D. in Cognitive Psychology from Carnegie
Mellon University in 2006. His research focus at N.C. A&T is on how humans in-
tegrate multiple sources of information He also conducts research projects in hu-
man memory and decision making.
Dr. Nader Vadiee leads the outreach activities at SIPI. He has more than 30 years
of experience in both academia and industry, and has conducted research and of-
fered courses in Electrical and Computer Engineering. He was selected by the
Carnegie Foundation as the 2009 New Mexico Professor of the Year, and he serves
on the New Mexico State Math and Science Council.
Dr. Brian Kelley is an Associate Professor of ECE at the UTSA. He is a leading
researcher on communication systems, 5G cellular, and cloud radio access com-
munications. He is the Associate Professor of ECE and Director of the Wireless
Next Generation Systems (WiNGS) Lab at UTSA.
Dr. Patrick Benavidez received his B.S. degree, M.S., and Ph.D. degrees in Electri-
cal Engineering from the University of Texas at San Antonio. He is currently a
post-doctoral fellow and an adjunct professor at UTSA.
Dr. Paul Rad received his B.S. and M.S. degrees from Sharif University of Tech-
nology in Computer Engineering. He received his Ph.D. in computer engineering
from the University of Texas at San Antonio. Currently, he is the Chief Research
Officer and Assistant Director of Open Cloud Institute (OCI) at UTSA.
SIAB Members
TECHLAV Center News Page 14
Edward Tunstel, Jr. Associate Director, Robotics United Technologies Research Center
Garry Roedler Lockheed Martin Fellow Engineering Outreach Program Manager Lockheed Martin Corporation
Roberto Horowitz James Fife Endowed Chair, Director, Partners for Advanced Transportation Technology (PATH), Professor of Mechani-cal Engineering University of California, Berkeley
Paul C. Hershey Raytheon Intelligence Infor-
mation and Services Senior En-
gineering Fellow
Daniel DeLaurentis Professor Department of Aerospace Engi-neering Purdue University
Alexander Eksir Vice-President
Supplier Quality
UTC Aerospace Systems
Russell L. (Rusty) Roberts Director Aerospace, Transportation & Advanced Systems Lab Georgia Tech Research Institute
Sandra Begay Technical Assistance Program Sandia National Laboratories’
tribal energy
Alton S. Wallace Senior Independent
Research Consultant
Julius Yellowhair Senior Member of the Technical Staff, Sandia National Laboratories
Mark Noakes Senior R&D Staff, Robotics and Remote Systems Oak Ridge National Laboratory
Bijan Sayyarrodsari Manager - Research & Develop-ment Center, Strategic Development, Rockwell Automation
student Spotlight
Wendell Bates What sets you apart as a student?
I believe that my experience working
in the industry prior to returning to
pursue my Ph.D. has helped me gain
perspective and appreciate the value of
education and the level of support that
I have received at the TECHLAV
Center. Working for BMW for four
and a half years in a management role
taught me invaluable skills dealing
with people and work/life balance. It is
very easy to get overwhelmed with the
workload and mental demands of a
job. Now I remind myself that the
journey should also be fun. I try to
bring that same mindset and energy
to the lab and make my colleagues
laugh or smile. I believe that we are
more productive when we’re happy.
The more productive we are as
individuals, the better we are as a
research group.
~ Wendell Bates ~
Wendell Bates received his Bachelor and Masters of Science in Electrical Engineering from Florida Agricultural and Mechanical University in 2008 and 2011, respectively. He then joined BMW Manufacturing, Inc. as a Quality Engineer and Electrical Systems Integration Engineer. In 2016, he joined North Carolina A&T State University to pursue his Ph.D. degree in Electrical Engineering. His research interests include Safe, Robust, and Reliable Control Systems, Hybrid Control Systems, Discrete Event Systems, and Robotics. He is also a member of the Autonomous Cooperative Control of Emergent Systems of Systems (ACCESS) Lab as well as the Testing, Evaluation and Control of Heterogeneous Large-scale systems of Autonomous Vehicles (TECHLAV) Center.
TECHLAV Center News Page 15
upcoming Events TECHLAV’s Fourth Annual Meeting is currently underway to be held from
June 25 until June 26, 2019 at UTSA. This meeting stands to capitalize on the major successes of the Third Annual Meeting and further TECHLAV’s efforts in autonomy, STEM involvement amongst mi-nority students and partnerships, groundbreaking research, and collaboration. The upcoming meeting is committed to excellence by presenting keynote speakers such as, Dr. Corey Schumacher, who serves as the Air Force Research Laboratory’s Chief Engineer for Autonomy in Motion and the Chief Technolo-gy Officer for the Autonomy Research Collaboration Network.
TECHLAV Center News Page 16
Seminar series guest speakers: Jan. 18, 2019 Ali Raz, Ph.D., Assistant Professor, Purdue University Feb. 1, 2019 Jeremy J. Dawkins, Ph.D., Assistant Professor, United States Naval Academy Feb. 15, 2019 Adrian Stoica, Ph.D., Senior Research Scientist, JPL Feb. 22, 2019 Tadilo Bogale, Ph.D., Assistant Professor, N.C. A&T Mar. 1, 2019 Madjid Fathi, Ph.D., Professor, University of Siegen, Germany Mar. 22, 2019 Amir Ajorlou, Ph.D. Research Scientist, MIT Mar. 29, 2019 Juan Morinelly, Ph.D., Research Scientist, Rockwell Automation April 5, 2019 Jim Gregory, Ph.D., Professor, Ohio State University April 12, 2019 Bijan Sayyarrodsari, Director of Advanced Analytics at Rockwell Automation April 26, 2019 Shaoshuai Mao, Ph.D. Assistant Professor, Purdue University May 3, 2019 Luca Carlone, Ph.D., Assistant Professor, MIT May 17, 2019 Dimitar Filev, Ph.D., Executive Technical leader at Ford Motor Company
TECHLAV Center News Page 17
sponsor Acknowledgement
The TECHLAV Center would like to acknowledge the Department of
Defense (DoD), the Office of the Secretary of Defense (OSD), and Air
Force Research Laboratory (AFRL).
The views and conclusions being discussed here are those of the
researchers and should not be interpreted as necessarily representing
the official policies or endorsements, either expressed or implied, of
Air Force Research Laboratory, OSD, or the U.S. Government.
Testing , Evaluation and Control of Heterogeneous Large-scale
Systems of Autonomous Vehicles (TECHLAV) Center
Fort IRC, Suite 221
1601 E. Market Street
Greensboro, NC 27411
Ph.: (336) 285-3271