Unmanned Vehicle Autonomy

Addressing the Limits to Unmanned Vehicle Autonomy Through Human-Computer Collaboration

Speaker: Andrew Clare

Science, Technology, and Society Program
Spring 2012 Colloquium Series
Date: Thursday April 12, 2012
Time: 3:30 – 5:00 p.m.
Location: Rodman Room
Thornton Hall
University of Virginia

In the future, teams of networked Unmanned Vehicles (UVs) will be utilized for military operations, border patrol, atmospheric research, forest firefighting, search and rescue, and cargo delivery among other uses. Recent advances in UV autonomy along with the development of optimization algorithms for dynamic scheduling of tasks for multiple UVs have enabled this future vision. The impending use of teams of humans and UVs in American airspace raises new ethical and legal issues. One key concern is that the uncertainty and time constraints inherent to command and control situations can cause poor performance by these automated systems. Thus, researchers at the MIT Humans and Automation Laboratory (http://halab.mit.edu) are developing mixed-initiative scheduling systems, where a human guides a computer algorithm in a collaborative process to solve the scheduling problem for the team of UVs. Examples from simulation-based experiments, indoor flight tests, and outdoor flight tests will show how such a human-computer collaborative system can best handle a realistic scenario with unknown variables, possibly inaccurate information, and dynamic environments. On-going research on this topic from Humans and Automation Lab will be described, including investigations into modeling both qualitative and quantitative aspects of how humans work with scheduling algorithms and teams of UVs. Finally, the impact of these research efforts on the relevant ethical and legal issues will considered.

Andrew Clare is a Ph.D. Candidate in the MIT Humans and Automation Laboratory in the Department of Aeronautics and Astronautics. He received his B.S. in Aerospace Engineering with Information Technology from MIT in 2008 and his M.S. in Aeronautics and Astronautics from MIT in 2010. He has held research internship positions with the U.S. Army Research Laboratories, General Electric Aviation, Aurora Flight Sciences, and the National Aerospace Laboratory in Amsterdam. He was the recipient of a National Defense Science and Engineering Graduate Fellowship (NDSEG) in 2008.

For more information on the STS colloquium series see:

http://www.sts.virginia.edu/stshome/tiki-index.php?page=Colloquium+series

Virginia Space Grant Consortium 2012-13 Graduate STEM Research Fellowship Program

The Virginia Space Grant Consortium (VSGC) Graduate Research STEM Fellowship Program provides fellowships of $5,000 in add-on support to graduate students to supplement and enhance basic research support. Applicants must be enrolled full-time in a program of study in science, technology, engineering, or math (STEM) and have a specific faculty-mentored research project that has NASA or aerospace relevance.

The purpose of the VSGC Graduate STEM Rresearch Fellowship program is to encourage talented individuals to pursue careers in the aerospace industry that support NASA’s mission. Awardees will be required to certify through their academic department that basic research support of at least $5,000 in matching funds is also being provided before funds are released. These Fellowships are intended to enhance basic graduate support such as graduate research assistantships, graduate teaching assistantships, and non-federal scholarships and fellowships.

Graduate Research Fellowships require that students participate in an active faculty-mentored research experience that aligns with the aerospace sector and meets NASA’s mission.   Awards are made annually and are renewable for one year for students making satisfactory academic and research progress. This is a competitive fellowship program and awards are based on merit recognizing high academic achievement and promise.

Eligibility

• Must be a U.S. citizen
• Must be enrolled as a full-time graduate student at one of the Virginia Space Grant Consortium member institutions
• May apply prior to entering or anytime during your graduate program
• Must be in a program of study demonstrating an interest in aerospace or supporting NASA’s mission
• Must have a grade point average of 3.0 on a 4.0 scale

Since an important purpose of this program is to increase the participation of underrepresented minorities, females, and persons with disabilities in aerospace-related careers, the VSGC especially encourages applications from these students.

http://www.vsgc.odu.edu/sf/gradfellow/index.shtml

APPLICATION Deadline February 10, 2012
(Letters of recommendation must be received by February 17, 2012)

Virginia Space Grant Consortium Scholarship

Dear University of Virginia SEAS Students,

The Virginia Space Grant Consortium (VSGC) will be holding an information session for upcoming student scholarship and fellowship opportunities. VSGC offers Undergraduate Research Scholarships of up to $8,500 and Graduate Research Fellowships of $5,000 (renewable for one year). The Undergraduate Scholarships can be combined with summer research or an intern position in a lab. These awards look great on your resume if you are applying for a job or to graduate school and SEAS students have a good track record of receiving awards each year, so I encourage to you to attend. Your studies or research needs to be in science, technology, engineering, or math (STEM) fields and some how related to NASA’s Mission (even if it is a stretch). The awards are only open to U.S. Citizens.

The information session will be in MEC 215 on Tuesday, October 18, 2011, 12:30 – 1:30 p.m and there will be free pizza and soft drinks. Please RSVP to jdotzaue@odu.edu

Thanks,

Christopher P. Goyne, Ph.D.
Research Associate Professor of Mech. & Aero. Engr.
Director, Aerospace Research Laboratory
Aerospace Research Laboratory
University of Virginia
570 Edgemont Rd
PO Box 400248
Charlottesville VA 22904-4248 USA

+1 434-982-5355 (office) 982-2037 (fax)
goyne@virginia.edu

Autonomous Quadcopters Work Together To Build Structures

Clay Dillow writes in Popular Science (1/19), “Whenever a new video emerges from UPenn’s GRASP lab (that’s General Robotics, Automation, Sensing and Perception), it’s usually awesome, and this one is no exception.” The video features a team of autonomous quad-rotor helicopters “working from a preset algorithm…constructing a cubic tower structure using specially designed parts that snap together via magnets when placed in the proper arrangement.” The quadcopters “can even judge the quality of their own construction, checking to make sure a piece is properly in place before moving on to the next segment.” Considering potential uses, Dillow writes, “Beyond the obvious applications in automated construction processes, swarms of construction ‘bots could be launched from naval vessels to autonomously construct shelters in disaster-stricken areas or to set up a forward operating base before live troops arrive in a combat zone.”

Reposted from the 1/20/11 ASEE First Bell

Reducing Jet Engine Emissions

U.Va. Aerospace Engineer Awarded $1.5 Million from Rolls-Royce and Commonwealth to Further Reduce Environmental Impact of Jet Engines

July 9, 2010 — Harsha Chelliah, a University of Virginia professor of mechanical and aerospace engineering, recently secured a four-year research grant to better understand and, it is hoped, eliminate soot particle formation produced by gas turbine engines. The grant, seeded by Rolls-Royce, the Commonwealth of Virginia and other partners, totals roughly $1.5 million.

Working from a lab in U.Va.’s decommissioned nuclear reactor facility, Chelliah and his graduate and undergraduate student research team have built two model reactors to study how to reduce the production of soot particles. The first experimental reactor operates at normal atmospheric pressure and has allowed the team to fine-tune experimental tools and methods.

Testing will soon begin on a second reactor that can burn jet fuel at 40 to 50 atmospheric pressures, which is equivalent to real jet-engine flight conditions. This undertaking fills a void in research at such high-pressure conditions and will help aerospace engine manufacturers meet anticipated U.S. Federal Aviation Administration regulations on soot particle emissions.

“While the aerospace industry has taken significant steps to reduce soot particles, there is still work to be done,” Chelliah said. “A major reduction or complete elimination of particulate matter formation by gas turbine engines is possible, with advanced diagnostic techniques being developed around the world.

“This research is expected to improve air quality near major airports and also reduce contrail formation in the upper atmosphere,” he added.

Both laboratory reactors are outfitted with advanced diagnostic equipment to capture soot particles illuminated by lasers. The process allows the research team to quantify the presence of particles at different burn conditions. Data produced will be used to create more reliable computational models for development of next-generation jet engines.

The project is one in an array of research activities born of a partnership among Rolls-Royce, the commonwealth, the U.Va. Engineering School and the Virginia Tech College of Engineering, formed in 2007 when Rolls-Royce announced it would build a new aerospace facility in Prince George County. Construction is now under way.

Jet-fueled, gas turbine engines produce vapor trails, or contrails – short for condensation trails – made up of droplets that form around nanoscale soot particles. From the naked eye, these are observed as jet trails often crisscrossing the sky, appearing as merely white lines in the atmosphere. According to some studies, contrails formed in the upper atmosphere may cause shifts in the global energy balance by trapping heat that radiates from the Earth, while soot particles emitted near airports during landing and take-off may have implications for human health.

The project funding supports several graduate and undergraduate students and research staff. The U.Va research team is working in collaboration with the Chemical Kinetics and Modeling Group at the National Institute of Standards and Technology, where Chelliah spent a year of sabbatical leave in 2008.

— by Zak Richards; reposted from UVa Today, July 12, 2010

In a Galaxy Far, Far Away…

What would happen if you pointed the Hubble Telescope at a seemingly “empty” part of the sky?  Find out at the Engineering.com web site!

Out of Many, One

Autonomous Aircraft Combine To Increase Stability, Power.

Popular Science (6/8, Dillow) reports, “Researchers at the ETH Zurich recognize that different tasks call for different aircraft, and with that in mind they’ve designed the Distributed Flight Array (DFA), a flying platform consisting of multiple small autonomous single rotor aircraft that can dock with one another to create a larger, more powerful aircraft.” Each “fixed [propeller] aircraft” that makes up the DFA has “its own sensors and flight control system,” and can “fly somewhat erratically.” However, “joined together they become a larger sensor-based flight platform, capable of maintaining level flight by rapidly sharing data between them.” The array “is a proof of concept” at this point, but “such a scheme could have a variety of applications, not least of which is the relatively straightforward yet sometimes difficult task of picking stuff up.”

Reposted from the June 8, 2010 ASEE First Bell

Natural Selection Among the Ions

Program Simulates Natural Selection To Improve Ion Engine Design.

New Scientist (5/24, Marks) magazine reported, “The life expectancy of a popular type of ion engine has been almost doubled using software that mimics the way natural selection evolves ever fitter designs.” Cody Farnell, a space flight engineer at the University of Colorado, used a “genetic algorithm” (GA) in order to change the girds typically used in ion engines, which typically last 2.8 years. The geometry and voltages of the grids were substituted for the genes the program typically simulates. If a configuration looked “promising, the ‘genetic material’ was subjected to further random changes,” ultimately leading to a configuration that could last 5.1 years if the simulation is correct. “The engine could be improved further, says Farnell, by evolving the other parts too.”

Reposted from ASEE First Bell for May 25, 2010

Scramjet Launch

X-51 Scramjet Will Launch Later This Month.

New Scientist (5/7, Courtland) reported, “In the last week of May, thousands of square miles of airspace above the Pacific Ocean will be cleared to make way for a skinny, shark-nosed aircraft called the X-51.” The scramjet will be dropped from a jet, where it will attempt to maintain hypersonic speeds for minutes. Boeing X-51 Program Manager Joe Vogel said, “No one has successfully flown a vehicle of this nature for more than a few seconds.” NASA’s X-43 was “the last US hypersonic scramjet to fly successfully…managing only 10 seconds of powered flight. Unlike its predecessor, the X-51′s engine uses novel active cooling systems and uses standard jet fuel.”

Reposted from the May 7, 2010 ASEE First Bell

A Fly on the Wall

Fixed-Wing Drone Lands Vertically On Walls.

Popular Science (4/27, Hsu) reports that researchers at Stanford University’s Biomimetics Laboratory have developed “a fixed-wing, non-transforming drone” that can land vertically on walls. “Their drone approaches the wall at full speed,” and “then pitches sharply upward to angle its belly toward the wall and slows its approach speed to just under 7 mph.” The drone uses carbon-fiber and balsa landing legs “tipped with steel spines” in order to make a vertical landing. “The researchers still face engineering challenges such as tuning the suspension system so that the drone doesn’t simply rebound upon landing approach.” They will be presenting “an update on their work at next month’s 2010 IEEE International Conference on Robotics and Automation in Anchorage, Alaska.”

Reposted from the April 27, 2010 ASEE First Bell