robotics and autonomous systems

Electrical and Computer Engineering

competitions

lighter-than-air robotics

Sea life defended the republic

Jennifer Pocock — Tue, 29 Apr 2025 15:34:12 +0000

Sea life defended the republic Jennifer Pocock Tue, 04/29/2025 - 11:34

The bioinspired blimps performed swimmingly.

Blimp Squad members Elijah Pointer and Kareem Zaharan (seniors in mechanical engineering) reach for a balloon captured in Narly's net as Faculty Adviser Daigo Shishika looks on. Photo by Jenn Pocock

Narly glided toward the ceiling of �鶹��Ƶ's Life Science and Engineering Building aviary like a shiny beluga whale with undulating manta-ray wings and a rudder tail. Thomas Girois, a senior in mechanical engineering and Narly’s team lead, flipped on the blimp’s autonomous goal-seeking program to show off its ability to target airborne objects. He hopes it will head toward the balloon that he wants to capture. But it made an unexpected turn and instead nosed the high bay windows, where it wouldn’t move despite its programmed back-up function.

After a moment of confusion, Girois laughed. “It’s outsmarted us again!”

Narly’s microcontroller and camera was coded to recognize and hunt down green and purple airborne objects—which is exactly what it did when it registered the glowing green Exit sign on the ceiling outside the window. Still, the test boded well for the upcoming competition.

Narly was one of three sea life–inspired Lighter Than Air (LTA) vehicles to compete on George �鶹��Ƶ’s Blimp Squad in the Defend the Republic (DTR) contest from April 21 – 25, 2025.

The Blimp Squad lines up with their devices, ready to go on the offense. Photo by Jenn Pocock

Flappy, another manta ray blimp, has a larger rudder tail that allows it to make pinpoint midair turns, and Knifey, inspired by the knifefish, has a long, undulating frill along its belly—and a fierce expression. These designs differ from their competitors’, which rely on propellers for propulsion and steering.

“Using propellers is easier,” said Daigo Shishika, assistant professor in the Department of Mechanical Engineering and one of the Blimp Squad’s faculty advisers. “But the big surface areas of our flapping wings are more efficient. Propeller-driven blimps need to carry more batteries, which are heavy and drain fast.”

Shishika noted that contestants lose points if they hurt another team’s blimp, which is always a risk with propellers.

Alexia De Costa (sophomore, computer science) follows under the fearsome Knifey, whose underbelly frill was inspired by the knifefish. Photo by Jenn Pocock

“We’re hoping to get more agility and a safer blimp,” he said.

Ningshi Yao, an assistant professor in the Department of Electrical and Computer Engineering and the squad’s other faculty adviser, also pointed out the blimps’ winning personalities. “Our flappies are pretty attractive robots,” she said. “Human-robot interaction is important. You want humans to like being around it!”

One DTR spectator, upon seeing Flappy fly for the first time, confirmed this theory. “It’s so cute!” she gasped. “I would die for Flappy.”

After hundreds of hours of testing, Narly, Flappy, and Knifey joined their two defense blimps—a zippy acrobatic wonder called Space Fish, and a high-scoring high-flier called Backbone Junior—to play a midair game of capturing balloons and releasing them through hoops to score goals.

Knifey's fierce expression intimidates the competition. Photo by:Jenn Pocock.

The team took third place out of six and scored the second-most points of all the teams.

“One highlight for me personally was that the flapping wing blimp achieved fully autonomous scoring for the first time since its first debut in the spring of 2023,” said Shishika. “It started as an exploratory agent that was primarily for research, but after two years and four competitions, Flappy finally contributed to the team by scoring 70 points.”

Narly also scored its first autonomous points.

While the team may not have taken first place, their sea-inspired research won the week.

Check out these shorts on our Youtube channel to see the vehicles and their designers in action.

In This Story

Daigo Shishika

Ningshi Yao

Electrical and Computer Engineering at George �鶹��Ƶ

Topics

Electrical and Computer Engineering

competitions

mechanical engineering

Applied computer science

�鶹��Ƶ's new MARC focuses on autonomy, robotics, and responsible AI

Nathan Kahl — Tue, 16 Apr 2024 14:30:37 +0000

�鶹��Ƶ's new MARC focuses on autonomy, robotics, and responsible AI Nathan Kahl Tue, 04/16/2024 - 10:30

Industry executives, campus leaders, faculty, donors and alumni attended the official opening of �鶹��Ƶ’s College of Engineering and Computing’s �鶹��Ƶ Autonomy and Robotics Center (MARC), but the star of the show was a Boston Dynamics quadruped. As the bright yellow doglike robot held one end of the green ribbon, �鶹��Ƶ President Gregory Washington, MARC co-directors Professors Missy Cummings and Jesse Kirkpatrick cut. Afterward, the quadruped roamed the event, opening doors for attendees and demonstrating its capabilities.

The event showcased �鶹��Ƶ’s strength in important, emerging fields. In his welcoming remarks, Dean Ken Ball said, “MARC is the focal point for our research in autonomy, robotics, and AI. �鶹��Ƶ truly is a pace-setter in these areas.”

Attendees explored the new facility in �鶹��Ƶ’s Research Hall on the Fairfax Campus. The space includes a 1,649-square-foot, two-story aviary for testing drones, areas for lab experiments, faculty offices, collaboration and study areas, and a student lounge.

�鶹��Ƶ President Gregory Washington and MARC co-directors Missy Cummings and Jesse Kirkpatrick are aided by a Boston Dynamics "dog" in cutting the ribbon to open the center officially. Photo by John Boal Photography.

During her remarks, Cummings addressed the evolving role of robots in society. “People don’t want autonomy to do their creative jobs,” she said. “They want to do their art, they want to write their stories…what they want is the robot maid. Once we hit world peak population in 55 years, we won’t have enough people to do the jobs we need. The dull, dirty, dangerous jobs are a great place for robots.”

Both Ball and Washington noted that in addition to being a leader in artificial intelligence (AI), the university is focusing on responsible AI, with a graduate certificate in that field launching in fall 2024.

“We’ve been at the forefront of a whole host of outcomes and technologies that have benefited this state and this region," said Washington. "Right here in engineering, we established the first-in-the-country cybersecurity engineering program, so there’s a real lineage of innovation here. �鶹��Ƶ will be the only Virginia university to join more than 200 of the nation’s leading AI stakeholders to participate in the Department of Commerce’s initiative to support the development of trustworthy and safe AI.”

“We do autonomy, we robotics, we do AI…but we do it responsibly and that’s in our core,” said Kirkpatrick. “It’s in our principles, policies, practice, and people. It’s across the research enterprise, and it’s in our teaching. The leaders of today, tomorrow, and the day after have to get this right because it’s so absolutely critical.”

Students demonstrate their claw game-like device, named Skygripper, to be used in the upcoming Defend the Republic competition. Photo by John Boal Photography.

Attendees explored the facility and talked to students about projects. Filling the MARC aviary for the event were student-designed “blimps.” These lighter-than-air creations will soon compete in Defend the Republic, an annual event in EagleBank Arena where the autonomous flying devices compete in a Harry Potter quidditch-like game to move a ball into a target hanging from the ceiling.

"Not just for engineering, MARC will be used by numerous colleges and departments across the university for many multidisciplinary projects. This is a collaborative space for our students to work,” said Ball. “And it’s really important they have the opportunity to work in close proximity to our faculty.”

MARC’s opening is another demonstration of �鶹��Ƶ’s strength in STEM, which attracts numerous corporate partners. Venkat Potapragada of 22nd Century Technologies attended the event and said, “I’m here because this is a strategic relationship for us and a key component of our overall growth strategy. We depend on talent, and �鶹��Ƶ will help us achieve our growth goals. In the D.C. region, �鶹��Ƶ is a name to be reckoned with.”

Enroll in the Certificate for Responsible AI

Learn More �鶹��Ƶ the MARC

Teaching humans to play safe

Nathan Kahl — Mon, 04 Mar 2024 21:02:44 +0000

Teaching humans to play safe Nathan Kahl Mon, 03/04/2024 - 16:02

Autonomous systems can be programmed to always make the logical, “best” decisions, given a set of circumstances. But what happens when human judgment and decision-making is introduced to a system? Xuan Wang, an assistant professor in �鶹��Ƶ’s Electrical and Computer Engineering Department, is asking this question as part of a recent $344,000 grant from the National Science Foundation.

Wang stressed that this research is particularly important given technologies on the horizon. “The operation of many real-world systems involves the co-existence of human and autonomous agents. Inadequate coordination among these agents can lead to significant performance degradation or safety risks.”

Wang is turning the idea of humans controlling machines on its head. “The key novelty of this research is, instead of thinking about how humans can program robots, we are thinking about the ways that the autonomous agents can impact humans," he says. "Assuming human response can’t be coded in the way we can control a robotic agent’s behavior, then how we can design the robot’s behavior so they're impacting human behavior in a way that is beneficial for the overall system?”

In one of Wang's simulations, a human and robot carry a cup of hot water, testing programmed guarantees so that when the cup spills, it only spills to the robot side of the board, protecting the human. Photo provided.

Because human agents, who are very diverse, use observations to see occurrences in the world around them and respond accordingly, traditional optimization approaches are less effective at predicting behavior. Wang says that he’ll use a framework relying on game theory, which assumes each agent has their own objective function, and that function is coupled with another agent’s decisions and actions. Then both human and autonomous agents ideally will optimize their overall behavior to coordinate across a whole system, creating a better output.

This human-response alignment mechanism is bidirectional, allowing for communication moving in opposite ways at the same time. For robots, they will investigate new approaches that allow them to adapt more intelligently to human behaviors with uncertainties; for humans, they will study how they can be incentivized during human-robot interaction so that human responses favor the efficiency and robustness of the entire system.

But how can systems—autonomous or controlled by humans—ever guarantee safety, say in the use of unpersoned vehicles?

Wang says, “When we are deriving safety criteria, there might be some uncertainties, so given the inputs of the system there will be an upper and lower bound that allows you to know what is the worst case that will happen. Given that, if all assumptions are satisfied, one can guarantee that there will be no crash.”

Wang and team are also working with the Army Research Lab to develop collaborative autonomous vehicles working in unknown environments, ensuring the vehicles can coordinate and gain advantage when there are potential threats in the environment.

In This Story

Xuan Wang

Topics

Department of Electrical and Computer Engineering

human-computer interaction

Artificial Intelligence

National Science Foundation

Research

CEC faculty research

Tech Talent Investment Pipeline (TTIP)

TTIP

Tech Talent Investment Program

AI's future discussed at roundtable

Rena Malai — Tue, 13 Jun 2023 19:36:06 +0000

AI's future discussed at roundtable Rena Malai Tue, 06/13/2023 - 15:36

In This Story

Missy Cummings

Just what is AI? 

This was one of the first questions posed by Missy Cummings, director of �鶹��Ƶ’s Autonomy and Robotics Center, at the Future of AI roundtable in June. 

“At the end of the day, AI can be a tool to help you in your job, but you need to understand both its strengths and weaknesses,” Cummings said. “That’s why �鶹��Ƶ will be offering a certificate and master’s degree focused on responsible AI, so that people across industry and government can learn how to manage the risks while promoting the benefits of AI.”

The invite-only roundtable, hosted by the College of Engineering and Computing, explored the issues, challenges, and solutions to think about as AI technologies rapidly evolve and change almost as soon as they are introduced. 

Missy Cummings leads the Future of AI roundtable at �鶹��Ƶ Square.

Cummings led the roundtable and highlighted AI areas like ChatGPT and autonomous driving systems that lack human reasoning skills. When human behavior, environment, and AI blind spots intersect, the resulting uncertainty contributes to fundamental limitations. 

“As humans, we do a great job of successfully filling in the blanks when there is imperfect information,” said Cummings. “If I quickly flash a picture of a stop sign, your brain automatically recognizes what it is, even if it’s not a clear image. But it’s arguable whether the vision system on a self -driving car can do the same.” 

She added that automation can fall apart at a critical threshold, such as when a car is supposed to stop at a stop sign that’s visible to the human eye, but maybe overlooked by an AI vision system. 

“As humans, we have much more experience driving cars,” she said. “Self-driving cars are still a new technology and while simulation testing can help identify problems, real-world testing for such non-deterministic systems that never reason the same way twice is critical.”

One of her favorite uses of ChatGPT is for her students to use it for grammar and spell checking when they write papers, but that’s the extent.  

“ChatGPT cannot reason under uncertainty. It does not think. It does not know. It can approximate human knowledge, but there is no actual thinking or knowledge,” said Cummings. “ChatGPT goes after the most probable image, or the most probable grouping of words.” 

In general, large language models like ChatGPT will use what the average person is saying on the internet. This means it could pick up extremist views, if it is currently something trending or popular online.    

It can also be a concern when it comes to diversity, she said. 

“If a company uses ChatGPT to write their mission statement, in another five years, everyone’s mission statement will be the same,” Cummings said. “Creative thoughts and authenticity are lost. It’s important to understand how far we push these models, and what the long-term ramifications could be.”

Roundtable attendees from local universities and tech companies left with a greater understanding of how AI can impact everyday life, in more ways than one.

Topics

Read the full article here.

Research

�鶹��Ƶ Square

Washington Post quotes Missy Cummings on Tesla's Autopilot

Rena Malai — Mon, 12 Jun 2023 17:11:55 +0000

Washington Post quotes Missy Cummings on Tesla's Autopilot Rena Malai Mon, 06/12/2023 - 13:11

In This Story

Missy Cummings

A Washington Post article titled '17 fatalities, 736 crashes: The shocking toll of Tesla's Autopilot' quotes Missy Cummings, director of �鶹��Ƶ's Autonomy and Robotics Center, on the troubling surge in Tesla crashes.

"Tesla is having more severe — and fatal — crashes than people in a normal data set,” she said in response to the figures analyzed by The Post.

One likely cause, she said, is the expanded rollout over the past year and a half of Full Self-Driving, which brings driver-assistance to city and residential streets.

“The fact that … anybody and everybody can have it. … Is it reasonable to expect that might be leading to increased accident rates? Sure, absolutely.”

She says in the article the number of fatalities compared to overall crashes was also a concern.

Topics

In the Media

�鶹��Ƶ wins Defend the Republic

Rena Malai — Mon, 22 May 2023 19:07:41 +0000

�鶹��Ƶ wins Defend the Republic Rena Malai Mon, 05/22/2023 - 15:07

In This Story

Cameron Nowzari

Ningshi Yao

Daigo Shishika

Patriot Pilots pose with their Lighter Than Air (LTA) creations. Photo by Ron Aira/Office of University Branding

Twice a year, university teams compete in the Lighter Than Air (LTA) Defend the Republic competition. In April, �鶹��Ƶ hosted the competition in EagleBank Arena and won. The winning �鶹��Ƶ team, Patriot Pilots, led by principal investigators Cameron Nowzari, Ningshi Yao, and Daigo Shishika, all from the College of Engineering and Computing, competed against six universities and about 50 LTA robotic agents, aka blimps.

“Even after three years, I’m still in awe of what happens at this competition,” said Nowzari, who has led the competition for the past three years, will be moving on from the annual competition and leaving things in the hands of Yao and Shishika. “My hope is that this competition will continue, and more schools will get involved.”

The competition is a mix of technology and creativity, and something students seem to have a lot of fun with, according to Yao.

“The strength of our winning team was in the ability to think creatively through building and design,” said Yao.

Without the severity that can sometimes accompany other types of flying agents, blimps are a lighter, gentler kind of airborne machine that appeals to many people.

“Their soft body and low operational speed make them more tangible and accessible than drones or planes,” said Shishika. “At the same time, they bring unique and exciting research challenges that our students with diverse background tackle together.”

The team will schedule a kickoff meeting during the summer, to plan for the next competition in November. The team welcomes students from all majors.

“We’re always actively recruiting for the team,” Nowzari said. “All that you need to bring is your passion and creativity.”

For information on how to join, email dshishik@gmu.edu or nyao4@gmu.edu. People can also visit blimp-robotics.org to learn how to build their own LTA blimp and encourage younger students to do the same.

Topics

competitions

EagleBank Arena

Strategic Direction

Blimps set to battle at EagleBank Arena

Martha Bushong — Mon, 17 Apr 2023 13:00:24 +0000

Blimps set to battle at EagleBank Arena Martha Bushong Mon, 04/17/2023 - 09:00

Learn �鶹��Ƶ the Robots

In the News

Patriot Profile: Dinesh Karri
April 24, 2023
�鶹��Ƶ Engineering students and faculty compete at 99 Luftballoons competition
January 8, 2021

In This Story

Cameron Nowzari

Ningshi Yao

Daigo Shishika

�鶹��Ƶ is hosting a competition of lighter than air (LTA) robotic agents—“blimps,” colloquially—squaring off in the heights of EagleBank Arena on Friday, April 28, from noon to 2 p.m.

The “Defend the Republic” competition features teams from six universities competing in a series of events where aerial robots attempt to move balls through hoops suspended from the ceiling, while other robots try to thwart them. The purpose of the game and the research behind it is to develop the ability of LTAs to respond quickly in various situations where such a vehicle, with its ability to stay aloft for much longer than traditional aerial devices, is ideal – disaster recovery, reconnaissance, delivering aid, and more.

At the Friday session, as many as 50 robots will be simultaneously in the air, with students and faculty available to answer questions about their robots and research.

Faculty members Cameron Nowzari, Ningshi Yao, and Daigo Shishika serve as principal investigators and lead the group.

Topics

lighter-than-air robotics

Access to Excellence podcast

Podcast: Missy Cummings: Artificial intelligence is artificial and not intelligent

Damian Cristodero — Wed, 25 Jan 2023 15:46:13 +0000

Podcast: Missy Cummings: Artificial intelligence is artificial and not intelligent Damian Cristodero Wed, 01/25/2023 - 10:46

Missy Cummings, one of the country’s first female fighter pilots and director of �鶹��Ƶ’s Center for Robotics, Autonomous Systems, and Translational AI, calls herself a tech futurist, charged with making tech work better and safer. In a conversation with �鶹��Ƶ President Gregory Washington, Cummings is unflinching in her critique of AI’s strengths, weaknesses, and shortcomings, as well as that of humans. There is a lot to like about AI, Cummings says, but she calls out bad tech where she sees it, including in the vision systems of self-driving cars and Tesla’s Autopilot.

Listen to this episode

Read the transcript

Intro (00:04):
Trailblazers in research, innovators in technology, and those who simply have a good story: all make up the fabric that is �鶹��Ƶ, where taking on the grand challenges that face our students, graduates, and higher education is our mission and our passion. Hosted by �鶹��Ƶ President Gregory Washington, this is the Access to Excellence podcast.

President Gregory Washington (00:26):
Missy Cummings is from a small town in Tennessee, where as she said few people finish college or even leave home. Her biggest challenge, she said, was finding the courage to go out into the unknown. All I can say is mission accomplished. Cummings, a professor in �鶹��Ƶ's Departments of Mechanical Engineering, Electrical and Computer Engineering, and Computer Science helped blaze a trail for women's equality in America's armed forces as a naval officer and as one of the Navy's first female fighter pilots. That's right: First female fighter pilots. That distinction came despite her facing discrimination and resentment from her male colleagues. She chronicled those events in her book "Hornet's Nest." Now the director of �鶹��Ƶ Center for Robotics Autonomous Systems and Translational AI, Cummings research interests include the application of artificial intelligence and safety critical systems, human systems engineering, and the social impact of technology. One of her first challenges at �鶹��Ƶ is to create a new educational program in the design and development of artificial intelligence. Cummings has asked the hard questions about the fundamentals of autonomous transportation while taking some jabs at bad technology, including Elon Musk, Tesla Autopilot, which we will discuss-- I'm an owner and so we can have a lot to discuss on this issue. She has been a guest of 60 Minutes, the Colbert Report, and the Daily Show with John Stewart. She also has a goal of hiking the entire Appalachian Trail. Dr. Cummings, welcome to the show and welcome to �鶹��Ƶ with the start of the spring 2023 semester.

Missy Cummings (02:21):
It is so good to be at both places. Thank you.

President Gregory Washington (02:24):
For those of you who may not know, Dr. Cummings and I have a long and storied history. I've actually tried to hire her on multiple occasions and I was ecstatic when we were able to get here, here with the help of Dean Ken Ball and others when we were able to get her here at George �鶹��Ƶ. So how far have you gotten on that Appalachian Trail?

Missy Cummings (02:45):
Oh my goodness. Well, I've only been doing it 20 years and you know, I have an attention span problem, so I can only go out for a few days at a time. So, you know, I'm about halfway. I still have a lot in Maine, New Hampshire up north. I'm pretty much done with the south though.

President Gregory Washington (03:01):
Nice. Well, you know, I am a hiker kind of guy in Virginia. There are some significant parts of that trail. I hear that there is the rollercoaster section. Have you done that section of the trail?

Missy Cummings (03:13):
Yeah, I'd knocked out the rollercoaster section a long time ago, but I will tell you, it was just, um, over COVID when, uh, down near Lynchburg, I went up in the wintertime and I took my Jeep and because I feel like I'm a fighter pilot, I could do anything and there's nothing I can't do. And I almost drove off the side of the mountain in my Jeep 'cause I hit a spot of ice that that was under the road surface and you couldn't see it. And I almost died. And, uh, whew. So Virginia is probably not the rollercoaster section that killed me, but over down by Lynchburg, that was dangerous.

President Gregory Washington (03:46):
There's just so much we could talk about and I want to jump in to a whole lot of it. So I wanna start with your time in the military because that was so defining for you in terms of your path now.

Missy Cummings (03:58):
Absolutely.

President Gregory Washington (03:58):
So you were in the military from 1988 through 99.

Missy Cummings (04:02):
That's correct.

President Gregory Washington (04:02):
So when did you start flying fighter jets?

Missy Cummings (04:05):
Well, I went through flight school--the first couple of years when you're a baby pilot, you, you fly propellers and then you fly a couple of different kind of jets. So I didn't become a full-fledged jet pilot until 1990, and it's at that time that I forwarded deployed to the Philippines.

President Gregory Washington (04:21):
Nice.

Missy Cummings (04:21):
And, and I was flying a four echoes back then.

President Gregory Washington (04:23):
Oh, you're flying a fours?

Missy Cummings (04:25):
Mm-hmm <affirmative>. Mm-hmm <affirmative>. I'm a real man.

President Gregory Washington (04:27):
<laugh>. So are you qualified to fly any other aircraft?

Missy Cummings (04:30):
Well, I flew F-18s as well. I then, when I became an official fighter pilot when I was in the Philippines, I was an aggressor pilot. So if you ever saw Top Gun 1, that's sort of what we did in the Philippines. We were pretending to be bad guys. And because I was a woman and at that time, women couldn't fly in combat, we just trained the men who were coming over to deploy in Iraq. And so we got to train the men to try to defend themselves. And it wasn't until a few years later that then the combat exclusion law was repealed. And that's when I, because I had already been doing the mission, I was one of the most qualified women to become a fighter pilot because I'd been doing those missions. That's when I rolled over to the F-18.

President Gregory Washington (05:09):
You know, <laugh>, I bring this up, I literally two weeks ago, just saw Top Gun 2 for the first time. I plan on seeing it again. Anything realistic about those Top Gun movies?

Missy Cummings (05:20):
The flying scenes are amazing, but it's not at all realistic. Right. I mean, the, the bottom line is these planes are so expensive that if you're getting that close to them, it's just too dangerous. You can't afford to lose the a hundred million dollar copy of the aircraft. And that's not even including weapons. And so there's a lot that goes on with the movies about the flying scenes that I think are not that realistic.

President Gregory Washington (05:41):
Oh so this, that whole one where he flies through the middle of the formation of the two aircraft, that probably wouldn't happen. That what you're saying?

Missy Cummings (05:47):
I mean, a lot of what happens at Top Gun you'd get kicked out for if you actually did that in real life. I, I. But they do a good job of capturing the spirit of what it means to be a fighter pilot.

President Gregory Washington (05:57):
No, I understand. I understand. You did a presentation a while back where you spoke about how, despite being an ultra trained and sophisticated pilot on takeoffs and landings in the F 18, you and all pilots are pretty much taken out of the equation as the plane flies itself at that point. You also wrote an extensive paper on the use of autonomous and automated weapons. Can you talk a little bit about that from an autonomy perspective? How much the person actually does and how much is involved in the actual technology of the plane itself?

Missy Cummings (06:28):
Yeah, I think these are really great questions, especially as autonomy and artificial intelligence continues to advance. So when I was flying, when I was transitioning from the A-4 to the F-18, one of the things that I found amazing was that when you were launched off the front of the aircraft carrier in an F-18, you had to show everyone on the carrier that you were not touching anything. You were not allowed to fly the plane. It was a computer program, it could fly itself off the front of the carrier just fine. The problem was that if you touched anything, you were likely gonna set up some pilot--induced oscillations that were not gonna be recoverable. And in fact, lots of people died this way. So this is why, you know, I felt that was really unnerving and that was really the beginning of me starting to wonder whether or not I should be doing something else, like going into academia. When I looked around the aircraft carrier and we weren't allowed to touch anything on takeoff because we would only screw it up. The planes always, always, always landed better than we did. Humans, we loved it--

President Gregory Washington (07:25):
Always?

Missy Cummings (07:26):
By, by that time, I mean the early days of automated landings, it was a bit sketchy. But by the late nineties and 2000, I mean, the planes are, computers can respond so much faster than we can as humans. We just can't process information as quickly on those kinds of jobs that the computer on the plane can.

President Gregory Washington (07:45):
Okay, hold on now. You are sitting in hundred millions of dollars worth of aircraft and you're on the aircraft carrier and they tell you don't touch nothing on takeoff.

Missy Cummings (07:58):
Yep.

President Gregory Washington (07:59):
The computer will handle everything.

Missy Cummings (08:00):
Yep.

President Gregory Washington (08:00):
And you're okay with that?

Missy Cummings (08:01):
Well, I didn't say it was okay with it, but it's what you do to stay alive. Right? I mean, it's very unnerving. It it is unnerving. And it's unnerving to watch the plane land itself always better than you can. You know, most of the really bad accidents on an aircraft carrier happen at 3:00 AM after a pilot's been out doing a mission, you're exhausted. You know, it's night, it's hard to see. And so that's a--

President Gregory Washington (08:24):
But if the plane is landing, why would you worry?

Missy Cummings (08:26):
Well, that's why you want it. Right? Because we have a lot fewer accidents now that the automation is at least assisting, if not doing it outright. I do think one of the things that people don't really get is don't be aghast at what I'm telling you about what's happening on a air, on a, on a military jet. It's happening to you every day when you fly commercial. Pilots only touch the stick for about three to seven minutes out of any flight, and it's on takeoff. Most of the time you're landing, you are being landed by a computer. And one of the reasons why the airlines locked it--

President Gregory Washington (08:57):
So, and even during the actual trajectory of the flight.

Missy Cummings (09:01):
Oh yeah. It's all automated.

President Gregory Washington (09:02):
They're not, they're not touching the stick.

Missy Cummings (09:03):
They're babysitting.

President Gregory Washington (09:04):
Well, let me ask you this. So you're flying, you hit turbulence, you bounce around for a few minutes and then you hear the ominous voice of the captain comes on and he says, or she says, we hit a little rough spot here and we're gonna glide down to smoother air. Is that the pilot taking over then?

Missy Cummings (09:19):
You know, it's kind of a hybrid. So the pilot's talking to air traffic control, getting a better spot. And then what they're doing is, like you're programming in your GPS, they're saying, okay, descend and maintain flight level 3 1 0 at an air speed of blahdi blah. Right. So they're just programming it in.

President Gregory Washington (09:34):
Okay. And then the airplane just,

Missy Cummings (09:36):
And the airplane, the airplane does it so much smoother. And indeed, one of the things that we've realized on landings and just flight in general is, first of all, if you let the computer fly, we save an amazing amount of fuel because pilots are just rougher on the controls. And so it's much more smooth when you let the automation do it. And it even saves on the tires. If You let the automation land the plane, they don't have to change out the tires as often.

President Gregory Washington (10:03):
That is so amazing. <laugh>. Wow.

Missy Cummings (10:06):
But, but let me tell you this, I would never get into a self-driving car that any of my students programmed. I'd fly in an aircraft a program, but I wouldn't get into a car.

President Gregory Washington (10:14):
We're gonna talk about that. I have never done any programming for aircraft relative to autonomy, but I have done a fair bit for automobiles, including a, just a couple of years ago on a project that we just completed on an autonomous dragster. And it, even going straight at high speeds is non-trivial. So I I I got a whole bunch of questions for you in that regard, but you mentioned that it was stressful, this whole idea of letting the plane land itself. I guess the human part of you wants to just take the stick and guide it down. And I assume at some point you have to take into account the fact that, you know, maybe you have a malfunction in a chip or the computer is not working properly and you have to land the aircraft. So I assume that at some point in time in training, you physically have to land on a carrier just so that you got the confidence that you can do it. Right?

Missy Cummings (11:07):
Well, and indeed airline pilots have to do the same thing. They have to land so many, so often to stay qualified, basically to stay up-to-date with your skillset. But indeed, you've hit on probably one of the biggest problems that we're facing in aviation right now. How much skill do you lose for the length of time that you let automation do it? And then how do we make sure that people keep their skillsets up even in the face of increasing automation. And the Asiana air crash, several years ago in San Francisco, there was a crash where there was a whole cockpit full of pilots and off-time pilots. And all of them missed the fact that, because the automation wasn't working that day they, you know, I think there were like five or more pilots in this cockpit and it still crashed and killed a lot of people in San Francisco. And that's because their skillset had eroded to the point that they didn't really even know how to fly the aircraft in the good old fashioned way. So I think there is a push and pull about either, and there's a lot of parallels to driving, like either the airplane can do it all the time with very high reliabilities, or you need to make sure that the human stays in the loop every so often. And now the FAA has to mandate the people get in and land every so often to keep that skillset up.

President Gregory Washington (12:23):
So talk to me about warfare. What is that like in an age of semi-autonomous systems? Is it closer to a video game?

Missy Cummings (12:31):
Oh, yeah.

President Gregory Washington (12:31):
Or, or is it closer to what we saw in Top Gun?

Missy Cummings (12:35):
I think it's kind of a mix. The reality is, is that there's a lot of automation that's finding its way into the cockpit. And one of the favorite stories I like to tell, and I told this in my book, is about a guy's call sign Spider. That's not really call sign. I changed it to protect the not-so-innocent. But when you're practicing missiles and you get the radar going, you actually would maneuver the airplane into an envelope. And if you got everything right, the envelope was right, the distance was right, the speed was right. And you would get these gigantic letters in your HUD shoot, shoot. You know, you'd pull the trigger and if you're on the test range, a missile would come off the rails and it would be at a static target. No problem. But there was a case where there was a squadron and they were deployed live over, you know, somewhere in the Middle East and Spider was coming back with his commanding officer.

Missy Cummings (13:22):
So that's be like me and you flying. And then they decided they had a little extra gas. And so they were gonna do a, a little one v. One top gun thing. They were gonna practice, pretend fight each other. But because they were coming back from a live area, they both had weapons on their plane.

President Gregory Washington (13:37):
So they had real weapons.

Missy Cummings (13:38):
Real weapons. But you can put the plane in simulate mode. So even if you have weapons and you pull the trigger, nothing happens, or you can leave it in live mode. And so when they went feet wet, which is when you go from the land to the water, they were supposed to go into simulated mode, but Spider got distracted and he forgot to push that button.

President Gregory Washington (13:56):
Oh.

Missy Cummings (13:57):
And so then they.

President Gregory Washington (13:57):
I know where this is going.

Missy Cummings (13:58):
So then they take a split, they come at each other and Spider's young, I mean the young guys usually have better reaction time. So he was able to get a bite onto the, his commanding officer, which means that he got to the, a good shot position first, and he lines up and he gets that amazing compelling shoot, shoot, shoot. And he shoots and a missile comes off the rail and then the planes tattletale on you. That's how you can't even lie anymore because as soon as a weapon leaves the planes, the video camera turns on. So it's like the police body cam is gonna turn on and make sure that it records everything bad you did. And so that you can actually see in this video, the missile go after his commanding officer, the commanding officer, because it was a heat seeking missile, he didn't have any of his systems on. He didn't even know this thing was in the air. And so you can see the missile gets like literally like inches from his tailpipe and then it just falls out of the air. It just didn't have enough juice to blow his commanding officer up. So the next thing you know, they have to come back to the carrier.

President Gregory Washington (14:51):
Wait, wait. So why, why did that happen? Was it just luck?

Missy Cummings (14:54):
Just luck. Just luck.

President Gregory Washington (14:55):
Just dumb luck.

Missy Cummings (14:56):
He took the shot on the very edge of the envelope and the missile just did not have enough gas to get there. So.

President Gregory Washington (15:01):
Oh my goodness.

Missy Cummings (15:01):
It's burning like a, and then it just kind of like a Bugs Bunny cartoon falls outta the sky. But then they have to come back land on the carrier. And so when you see a fighter jet and one missile's on one side of the plane and there's a missing one on the other one, it's not like you could say, uh, I don't know what happened. So, so he had to fess up.

President Gregory Washington (15:18):
What happened to him?

Missy Cummings (15:19):
In the old days, he probably would've been kicked out. But I think that they realized, that was the big time of the military was like, wait a minute, maybe we shouldn't have these shoot cues that are so compelling because it's making people respond in a video game-like environment instead of taking the time to actually think about, is this something that I need to be doing? And so things have changed since then, but it's a good story to indicate humans under stress and battle even, that wasn't even a real battle. Right. He's just excited and he was gonna be able to quote unquote, you know, fake kill his commanding officer. And he almost did.

President Gregory Washington (15:50):
He almost real killed him.

Missy Cummings (15:51):
That's right.

President Gregory Washington (15:52):
So we hear about these patriot missile systems now in the public. They're hearing a lot about this autonomous aerial vehicles. But the reality is, in a non-military sense, the deployment of UAVs is probably a thousand, 10,000 to one relative to what we're seeing in the military. I mean, these things are being used all over the place. Are you doing any work or working on any applications, uh, that are non-military?

Missy Cummings (16:17):
Oh, I haven't done dedicated military drone work for a long time. There was a point in history where you could see the tide turn from anti-drone sentiments because of military to pro-drone sentiments. That year was 2013. And I had been working really hard to try to socialize the idea that drones would not be just a military platform, that they had a lot of good. And I was trying to socialize in America the idea that these would be cargo planes one day. And so I got invited to go on the Daily Show with John Stewart. I recommend everyone go look at this clip 'cause it's hilarious. 'cause he is going after me for basically being part of the war machine. And I'm trying to explain to him that these are going to be delivery aircraft in the future. You know? And he and I had a good repartee of going back and forth about were these things really killer robots or are there some good to this.

Missy Cummings (17:14):
And that that was 2013. And what's amazing is 10 years later it's a done deal. Right, right. And then many years--

President Gregory Washington (17:19):
They're everywhere.

Missy Cummings (17:20):
That's right. Everywhere. And I've been to Timbo, Africa using drones to follow elephants around. They have a really hard time of keeping track of their elephants and making sure the poachers aren't getting them. So we could use drones for those applications. And then recently I finished a project sponsored by the National Science Foundation, looking at how we defend against drones in prisons, right. Because now one of the problems that we have is that drones are putting contraband into prison yards. And so now we were trying to come up using some artificial intelligence with ways to defend against the drones. And I think what was interesting--

President Gregory Washington (17:57):
There's a way, it's called a shotgun

Missy Cummings (17:58):
<laugh>

President Gregory Washington (17:59):
<laugh>.

Missy Cummings (18:01):
Yes. It turns out, do you know you're not allowed to do that? The FAA says don't do it. Right.

President Gregory Washington (18:05):
So you can't shoot.

Missy Cummings (18:06):
You can't, even if there's one hovering over your house, technically it's illegal for you to shoot a drone hovering over your house.

President Gregory Washington (18:12):
Really? I guess that's right. Because the property above your home, is it?

Missy Cummings (18:17):
That's correct.

President Gregory Washington (18:18):
25 feet and higher,

Missy Cummings (18:20):
No, it's like one inch.

President Gregory Washington (18:21):
But that is not your property, actually. It's

Missy Cummings (18:23):
Not, it's not your property. And fa a doesn't want you shooting things because they don't know where the drone is gonna go.

President Gregory Washington (18:28):
Oh, well, not only that, you shoot and then if you miss the projectile lands somewhere, it does come back down.

Missy Cummings (18:35):
So, so my advice is, if you're worried about that, just get a big old light and put it on top of your house and direct they, they can't, that'll totally screw the system. So there's lots of passive ways that we can defend against these things.

President Gregory Washington (18:45):
So you, you would come up with the, the technical way to defend against them?

Missy Cummings (18:49):
Oh yeah. Oh, oh yeah.

President Gregory Washington (18:49):
As opposed to the, you know, the good old American way. <laugh>, shoot it down. Shoot it down.

Missy Cummings (18:54):
Okay. Well there's part of me that really wants to do that. Like, I didn't say you couldn't use a slingshot.

President Gregory Washington (18:58):
<laugh> exactly. Okay, I hear you. So let's switch gears a little bit. There's a lot of talk these days about Elon Musk with the whole Twitter issue and him purchasing Twitter, but that has spilled over to Tesla. I'm a both a Tesla owner and a Tesla stock owner, have been for quite some time and have seen, uh, <laugh> the value of my Tesla shares decrease dramatically over the last year. So talk to me about your challenges with Elon Musk.

Missy Cummings (19:29):
So, I mean, it's hard for me to say that I have a war going with one of the richest men on the planet, right. Because it's only his perception that that's the case. I'm a big tech futurist.

President Gregory Washington (19:41):
Right.

Missy Cummings (19:42):
Uh, that's my job is to try to make tech work. It's not to stop tech. It's to help it get better. And I've been a big fan of SpaceX for a long time. As far as Teslas, I think they're great cars. I think that certainly they're very crash worthy. After you saw that Tesla go down that cliff and everybody survived, I'm like, you know, that thing has a good cage. That, that is a solid car.

President Gregory Washington (20:04):
Yeah. Because it's unibody construction.

Missy Cummings (20:06):
That's right. So I am not anti-Tesla, but like I will tell you, and there's many people in the drone world that know this, and in the driving world and in the AR/VR world--augmented reality, virtual reality world: I just really hate bad tech. And if you've got some bad tech that's really dangerous, I'm gonna call you out on it because that is my job to make safe good tech. And the problem, and I hope that you're listening to me, is,

President Gregory Washington (20:34):
Oh, actually I hope he's not listening, but keep going. <laugh>

Missy Cummings (20:36):
Do not drive your Teslas on autopilot or full self-driving without paying full and absolute attention, and keeping your hands on the wheel.

President Gregory Washington (20:45):
So two things I I I will highlight to you: I can't help it, I'm an engineer, right. I put mine in, uh, auto drive mode all the time. And I can tell you the pluses and minuses to it. Technology is not quite yet ready for primetime without question. You know, and sometimes small artifacts, some of which I don't even see, cause the autopilot to stop working. What happens in almost every single case is the vehicle just abruptly slows down. So, and it's a scary thing when you're driving 70, 75 miles an hour on the highway and the thing just hits the brakes and it slows down dramatically. Maybe it saw a shadow, you know, you don't know what it saw, but it saw something that triggered a response. And I tell you, it's probably happened to me a dozen times. That being said, it is a remarkable technology to use when we're doing the kinds of things we're doing in our cars.

President Gregory Washington (21:41):
You know, oh boy, I'm giving a whole, it's gonna be funny, you know, but you're in the car and you're driving and he's like, oh man, I gotta blow my nose. Okay. Engage autopilot, things driving on its own. I can reach down into my glove box or into the center console, pull out a tissue, blow my nose, and put it back and it's all cool. I do that, no problem, right. And feel very, very comfortable doing that. Or if I'm coming home long time at work, little tired and need that extra hand: it's not so I fall asleep, I'm still driving, got my hands on the steering wheel, 10 and 2, so I'm still there, but I turn it on just so that I won't drift. That actually works quite well for me. So I do think there are uses, right now, as an assistant--

Missy Cummings (22:27):
I, I agree.

President Gregory Washington (22:28):
to the actual driver, right? We're not at the point where we can totally turn it over to the computer. And this is the thing that's amazing to me. You won't turn the car over, but you put lead <laugh> a hundred million dollar bird out of the sky onto a strip of concrete <laugh>. But you, you feel me here.

Missy Cummings (22:50):
Oh yeah. That's because I know how the sausage is made and I helped make that sausage. And so I see the mistakes that are made and I see the problems in the system.

President Gregory Washington (23:00):
I actually had a number of conversations, not just with engineers from Tesla, but also Zoox, which is an, a company was built to build autonomous vehicles. Say, Hey, well what's the problem? You guys are working on this technology every day. What are you struggling with? Why don't we have it and have it now? What are your thoughts? What do you think?

Missy Cummings (23:19):
So, you know, you hit on one of the issues. I just finished about a year and a quarter with the National Highway Traffic Safety Administration as the senior safety advisor. And for the last year or so, I've been looking at all the accident reports of any car, including Teslas who had a crash while they were on automation. And so this phantom braking issue that you described, where the car sees something and then decides to dramatically decelerate. So that is not just a Tesla problem, we see it in many other kinds of autonomous vehicles, including ADAS Equip, that's the driving assist systems, and also just the self-driving systems. So we have not yet gotten to the point where computer vision systems, they're just not reliable enough to be able to "see the world in the way that we do."

Missy Cummings (24:09):
And we don't know, is it shadows? You know, we've done some testing with Teslas in my own lab where we have, we can see a statistical correlation with the sun going behind clouds for, even that is enough potentially to trigger a problem with the vision system. So these systems are still really brittle. And I'm not saying we'll never get there, but we're still working out some very, very basic problems. That's just one of many problems. And that, that's the tech problem. But I loved you describing your reaching over to the glove box because I am here to tell you, you guys heard it from me first, that if President Washington ends up at a Tesla accident, it's gonna be because of the accidental steering nudge bump. So one of the things that we've seen in accident mode that we see people do is in these cars, and it's not just Teslas, there's also Blue Cruise and Super Cruise. People are so confident in the systems that what they do is they drop something on the floor, they need to reach around the back of the seat and pick something in the back seat up. Or they need just to get something outta the glove box and they reach across and their shoulder

President Gregory Washington (25:09):
Bumps the steering wheel.

Missy Cummings (25:09):
Just bumps the steering wheel And that, and sometimes, depending on the car is, and depending on the speed that you're going but lemme tell you something else I found, but

President Gregory Washington (25:17):
But the Tesla will go, it'll give you an audible signal, boom boom. And then that'll let you know that it's disengaging autopilot. And you do have some time to adjust sometimes, you know.

Missy Cummings (25:27):
But sometimes you don't. Sometimes you do it at the exact wrong time.

President Gregory Washington (25:30):
No, I get it. I get it. And I, and I get that that could be a problem. Here's the deal. You mentioned something earlier that I thought was really, really interesting. You said that in fighter aircraft, as automation became more and more prevalent and the technology became better and better, you actually started to see accidents decline. I believe you're gonna see a similar thing now. You're seeing accidents go up now, but you gotta correlate that with the fact that there are more and more of these vehicles in the market. I think a comprehensive study may show that the whole host of technologies that are in vehicles now, right? The lane departure warning, the auto-steer that pulls you back, we're probably seeing an overall decrease in the number of minor accidents that would've occurred 'cause you sideswiped somebody or you're coming up to a traffic stop and they would be looking at a text message and run right into the back of the car. I got rear ended that way. Nowadays, vehicles do catch you from doing that. They will stop the vehicle before or at least let you know with blaring signals that an accident is imminent if you don't do something. And so we should start to see, as the vehicle becomes, as the computers become more and more prevalent in how we drive, we should start to see the number of accidents going down, which is gonna have a dramatic effect on insurance companies because they make their money when there are accidents. <laugh>.

Missy Cummings (26:55):
Yeah. Right. Well, I I tell you, they're not too worried right now. Automation, basically, there's two different kinds in cars. There's the safety automated: auto emergency braking, the frontal collision warning. Right? These kinds of safety devices.

President Gregory Washington (27:08):
But those gotta be working.

Missy Cummings (27:08):
They and they are working. And indeed we can see that decrease that you're describing that is happening. But the Teslas, Super Cruise, Blue Cruise, these are convenience features that do lateral and longitudinal control for you. Right. They're doing the acceleration for you and they're doing steering on, right? So the jury is very much out. And there are a lot of scientists, myself, there's some other people at George �鶹��Ƶ looking at this. That jury is out. And I will tell you that having come from NHTSA, I did the analysis myself on all this crash data we have. And I will tell you that if you are in an accident in a car with these convenience features, right, you are statistically more likely to be seriously injured or killed.

President Gregory Washington (27:53):
Really?

Missy Cummings (27:54):
And there's one reason, one really big reason, there's a lot of little reasons...

President Gregory Washington (27:57):
'cause, because you're overdependent?

Missy Cummings (27:59):
That is probably, but there's actually one clear measurable problem.

President Gregory Washington (28:03):
What's that?

Missy Cummings (28:04):
You're speeding. So this is,

President Gregory Washington (28:06):
And that's 'cause of overdependence.

Missy Cummings (28:07):
Well, right. So this is the kind of interaction that we're seeing is that people become so reliant and they love their vehicles and they...President Washington is loving and trusting his vehicle so much that, you know, I'm just gonna go nine miles over the speed limit.

President Gregory Washington (28:20):
Yeah, yeah. No, look, first of all, without incriminating myself too much <laugh>, you hit the nail exactly on the head. <laugh> my speed goes up because I got that technology with me, without question. So now you're making me rethink. Maybe I need to tone the speed down.

Missy Cummings (28:34):
You do need to tone the speed down.

President Gregory Washington (28:36):
And so I will, I will definitely do that. But that brings me to another question because there's a YouTube video out there, and I know you've seen it, with the title, "Missy Cummings wants to destroy Tesla." <Laugh> True statement, overstatement, or what?

Missy Cummings (28:50):
No, of course it's an overstatement. I want Tesla to survive. I mean, my favorite thing about Tesla is the fact that it doesn't have a dealer model. Like if you wanna be, you go with any woman to try to buy a car and you will realize how much women hate the dealer models. Right? So we would love...

President Gregory Washington (29:07):
It ain't just women. Hello. Hope they look like me. You, you, you know what's I'm saying? Hey, look, I, I, I've been there. And um, you're absolutely right. It was so easy. You can pull up your computer right now and within 15 minutes I can buy a Tesla.

Missy Cummings (29:22):
For a lot cheaper these days.

President Gregory Washington (29:23):
Yeah, that's exactly right. Because of the "Missy Cummings wants to destroy Tesla" video. But <laugh>.

Missy Cummings (29:29):
I wish I, I wish I had that kind of power, but I will, I will tell you, look, I love the car itself is great. The model behind the car in terms of the deal, no dealer model.

President Gregory Washington (29:39):
Oh, that's fantastic.

Missy Cummings (29:40):
The picture of like, I, there are so many good things to love about Tesla, but I think that Tesla, they were first out to try to do something brave and innovative. And I get that. But now one of the other things I call it, you know, your mom always said, if you see your friends jump off a cliff, are you gonna go jump up a cliff? So now Tesla had some questionable design decisions about letting people be hands free, but now all the other car companies are modeling after Tesla.

President Gregory Washington (30:06):
Right. I hear you.

Missy Cummings (30:06):
And I do not think we should allow that. I think no car, not Tesla, not Ford, not GM, no car should in any driver assist, should allow you to be hands-free. And that is a very unpopular opinion. But unfortunately the Teslarati wants to try to blow that up into something like Missy Cummings is coming for your autopilot.

President Gregory Washington (30:24):
Yeah, no, I get it. Now look, if there could be wholesale adoption on the manner in which you buy and sell Teslas, that I tell you would be a game changer. The reality is it was so easy for me to buy my car. It literally took me about 15 minutes.

Missy Cummings (30:43):
Yeah. I'll do free advertising for them for that. Like I love that feature of their car. But I think the whole Teslarati thinking that I'm out to get them, it kind of points to the bigger problem. There's two problems it points to: number one, women in tech, women who assert themselves in tech. You know, it's funny, we, we talked about the fighter pilot thing was I discriminated against as a fighter pilot? Yeah. But I'll tell you what's shocking to me is the fact that I was a fighter pilot, carry a PhD, have been a tenured MIT professor, have done all of these things that the Teslarati and other tech bros hate the fact that I'm asserting myself and that I'm a broad. I'm a pushy broad, trying to push my opinion that is not favorable to their stock price. Right. So that's one issue.

President Gregory Washington (31:27):
That's what you think it is.

Missy Cummings (31:28):
Yeah, but I think also it points to the devisive nature of this country. Like I have a very, I think, balanced view of Tesla. It's a great car except for this bad autopilot that when you have your hands free or it basically promotes you into complacency. So I can like the car, but not like a feature. But that ability to have a balanced view towards really any person, politically, to a technology. Like you're either with me or against me. You know, kind of That's how people,

President Gregory Washington (31:59):
It's all, you're either in it, you're all in, or you're, or you're all out. That's right. I get it. I get it. A hundred percent. This is interesting. In the last few minutes I have, I wanna steer us more closer to your research and what you're doing or what you will be doing here at �鶹��Ƶ. The real advance in all of this is intelligence, right? We are bringing more and more intelligence to the systems, right? Whether it is classic neural networks with back propagation or Kohonen networks and the like, or deep learning, or it's just the idea of bringing expert modeling and systems into code, right? Where you take into account hundreds and thousands of variables in terms of decision making. The reality is, is that systems are getting more intelligent and you stand at the forefront of this. And so talk to us a little bit about the degree program you're putting in place and how do you see that fitting in to everything that you've learned up to this point?

Missy Cummings (32:58):
Yeah, these are great questions. You say that intelligence is advancing. And I will tell you, an approximation of intelligence is advancing. Understood. So artificial intelligence is artificial and not intelligent. And if you've heard about GPT, the large language models.

President Gregory Washington (33:13):
Yeah, yeah. ChatGPT.

Missy Cummings (33:14):
These are things are dangerous because they're good enough to approximate language. But if you actually pay attention, you can see very quickly how wrong and dangerous disinformation coming from something like chat GPT could be. But I've spent a lot of time, obviously in the aviation world now in the surface transportation world. I've spent some time in the medical world and looking at these large language models. And the one common theme across all of these are intelligence technologies are advancing so rapidly. What we're not doing is keeping up with allowing people to get educated in how to think about the design frameworks behind when should you have these systems? Why should you have these systems? What requirements are they really meeting? And then how should I test these systems to make sure that they're sufficient?

Missy Cummings (34:01):
And this whole idea of the design life cycle around AI, it's new thinking. Like people think, oh yeah, we know how to design systems. We've got agile system development. Well it turns out for safety critical technologies, maybe your testing framework needs to be a little different. Maybe you need to do different kinds of component testing. And guess what? Digital twinning, like I'm so sick of hearing digital twins because you can digital twin AI all you want, but garbage in, garbage out. The only way, if you're ever gonna know if your Tesla is actually going to not hit children, and this is a big debate going on right now in the Tesla community, is you do have to put it on the road and you do have to put it in various tests, real tests, not fake tests. Not FSD full self-driving tests. Like really principle tests that are answering and research question. And, and I think companies are reluctant to do this because it's expensive. It takes time and effort that maybe they wanna spend other money on. But

President Gregory Washington (34:57):
But the other thing is that they could fail. And when you fail the,

Missy Cummings (35:02):
It's more development cost.

President Gregory Washington (35:03):
Well, not just that the results are catastrophic. Right? I remember looking at the Tesla stock price when the first Tesla fire hit the news and you just watched the share price drop. That's somebody's livelihood. And reality is these are complex systems. Complex systems will fail, right? You have 'em in automobiles, you have 'em in rockets, you 'em in airplanes. You have 'em in fighter aircraft, right? There's been failures. Failure is a part of the process. You hope that you can put it in the context where there's not loss of of life. The reality is that these things do happen.

Missy Cummings (35:40):
Yes. And, and I agree with Henry Petrosky, he's a famous Duke professor who says to, to fail is just a core component of engineering. I'm all about that, right? But I think with artificial intelligence, one of the problems that we're seeing is that there just really aren't testing paradigms to try to at least figure out how to mitigate risk.

President Gregory Washington (35:59):
No, no, I get it. I think it's a little more, and I don't want to use the word nefarious 'cause I don't think people are trying to do harm. I think the challenge is a little different in that nowadays we don't know when we're interacting with AI technology and when we're not. Right? It's not ubiquitous yet, but it is far more intrusive in our everyday lives than we actually realize. And so you can be interacting with your vehicle, not a Tesla, but we own a BMW as another vehicle. Right? You could be interacting with that vehicle and there could be aspects of artificial intelligence handling some systems and you actually have no idea, right? People are dealing with ChatGPT and there, they're being told that they're dealing with artificial intelligence, but they're dealing with a whole host of technologies on their computers as they go to websites and as they frequent the internet on a day-to-day basis where they're not told and they're interacting with something, thinking that they might be interacting with a human and they're actually interacting with a bot, right? You would handle things differently if you knew it was a bot relative to a human. And so we need guardrails.

Missy Cummings (37:07):
And that's exactly what we're gonna teach you at �鶹��Ƶ.

President Gregory Washington (37:10):
Outstanding.

Missy Cummings (37:11):
We're gonna teach you how to build them, how to set systems up to design them, how to interpret them, how to recognize when you need guardrails. So this is one of the things I think that �鶹��Ƶ is just has such a rich field to pull from. There's many, many government agencies here. There's lots of top talent faculty here. Lots of really motivated students who are gonna work in all aspects of industry. We've got healthcare, we've got DOD, DHS, transportation industry. So I'm really looking to build a strong cohort of people who can recognize, do I need guardrails? What kind of guardrails? And how do I maintain those guardrails over time?

President Gregory Washington (37:52):
�鶹��Ƶ is constructing the Fuse building on its �鶹��Ƶ Square campus in Arlington. As you know, the building will house research labs, corporate innovation centers, incubators and accelerators. How does that interdisciplinary model fit into your research?

Missy Cummings (38:08):
Well, I'm hoping personally to teach classes in that building and actually have an offshoot of my lab out there. Because with all this work that we're doing with government agencies on safe, secure, trustworthy AI, we anticipate offering research and lab-based classes out there. So it's critical to my research and critical to the overall interdisciplinary nature of AI in general.

President Gregory Washington (38:33):
Well look, I am looking forward to what you're gonna do in Fuse. I think it's going to be fantastic. Just talk a little bit about how academia can be the agent that educates industry and government employees to actually ask the right questions about AI's performance, its weaknesses, its strengths, and its shortcomings.

Missy Cummings (38:54):
Yeah. So these are great questions. I think first and foremost we have to recognize...to start looking at the assumptions in the design and construction of AI. So I think what a lot of people miss is they think that AI is this great computational tool, one plus one is two. And so how can you argue with the math that's coming out of AI, for example? Well, it turns out that there's a lot of subjective work. And I personally have done research and I'm continuing to do this research that looks at the assumptions that modelers make. So when you're engineering, your computer scientist develops an algorithm, for example. They actually make a lot of guesses about how to initialize certain parameters inside the algorithm. How do I set some hyper parameters? And they don't really understand that the way that they set up the problem can actually cause the model to have very different outcomes as opposed to maybe another engineer who sets up a problem.

Missy Cummings (39:50):
So one of the research projects that I'm working on now that's gonna continue probably out at the Arlington campus is looking at data labeling. So it turns out, after spending some time with Amazon, I learned just how much data labeling is happening in offshore places like India and around the world. And lots of companies are using them. And then the question is, if you have people labeling images for eight hours a day, is that labeling just as good in their eighth hour as it is in the first hour? And one of the things that we're looking at my research right now is how does sloppy labeling not wrong labeling? So it's not wrong, people weren't circling the wrong image, the component of the image, but they were very sloppy. And then when you run that through a convolutional neural net, how much of the sloppiness and the data labeling shows up in the quality of the outcomes? Turns out it's pretty significant. And so.

President Gregory Washington (40:46):
That's part of the data set that's gonna be part of the model.

Missy Cummings (40:49):
That's right. So I really want to help people focus on knowing when, where, why, and how to ask those questions about the underpinnings of AI. Is there an assumption that was made in the development of this AI that could have a downstream effect? And not that you then shouldn't use the AI with that downstream problem, but at least you know that then there is potentially on the, on the downstream side that you have to maybe not trust the outcomes as much as you would if you had better quality data going into it.

President Gregory Washington (41:21):
Outstanding. I'm so looking forward to what we're gonna be able to give to the community, especially as this field continues to grow and as it continues to have an impact on taxpayers supported dollars. Right. The investment that the country is making in these technologies, you need to have an understanding of when to use 'em, when not to use them, and when to be cautious about their use.

Missy Cummings (41:46):
Right, and when people make big claims, I would like to give people tool sets to be able to evaluate those claims for themselves.

President Gregory Washington (41:54):
Outstanding. So I get to ask you a controversial question.

Missy Cummings (41:58):
Oh, what? We haven't been <laugh>.

President Gregory Washington (41:59):
Really, a really controversial one.

Missy Cummings (42:01):
Okay.

President Gregory Washington (42:02):
How long, in your opinion, before we actually see full self-driving vehicles? Well,

Missy Cummings (42:06):
I'm just gonna need a definition from you first. Do you mean like

President Gregory Washington (42:10):
<laughs> I mean you get in the vehicle and there's a steering wheel you can take over. So the whole concept from Zoox where you, there's no steering wheel, you just get in and ride: I'm not talking about that. But vehicles that are full self-driving where you have the option to say you push a button and the car just takes over.

Missy Cummings (42:28):
And you get in the backseat and go to sleep if you want and it'll take you to Vegas.

President Gregory Washington (42:31):
Well I'm not talking about that <laugh>. But you actually have the ability to do that: The technology will be sophisticated enough that you could indeed go in the backseat and go to sleep. When do you think that'll happen? Now not, I'm not saying that we will ever get to a point where the community allows that to happen, but when the technology is mature enough to happen?

Missy Cummings (42:51):
We're not even close.

President Gregory Washington (42:52):
We're not even close even. So how many...

Missy Cummings (42:54):
You know, I'm gonna pull a typical academic response. Oh, 10 to 15 years. 'cause that's the secret academic speak for, we don't know <laugh>, we, we have no idea. Right.

President Gregory Washington (43:03):
It's far enough away that where you can't get called on it.

Missy Cummings (43:06):
But yeah. Right. So I think that that we will see in the short term, there's been a lot of success. I mean you see it on George �鶹��Ƶ's campus with Starship little grocery delivery. So companies like Neuro who have the bigger vehicles are on the road that the purpose-built. Yeah. I think that they--

President Gregory Washington (43:22):
Those are working!

Missy Cummings (43:23):
Those are working. And we, I think that there's a real legitimate profit building.

President Gregory Washington (43:28):
What is it? TU has the trucks out there.

Missy Cummings (43:31):
So I think that small last mile delivery is probably where we'll see that first happen. You know, it just like Waymo is struggling still. Cruise is under investigation by the National Highway Traffic Safety Administration. I mean, all signs are like they're making incremental progress. But if you're asking me, should I go ahead and start investing in self-driving cars because they're gonna start turning a profit next year, I don't know when that year is gonna be.

President Gregory Washington (43:58):
Here's the last controversial question. What do you say to people who worry that automation will be taking away even more jobs from people? 'Cause we know they have taken away some jobs, we know where this is going. Talk about that a little bit.

Missy Cummings (44:13):
Yeah, I don't think we know where it's going. I think we think we know where it's going because we hear, you know, the media's job is to kind of get you to click more. So the headlines are always bad on this front, but I've been predicting this correctly for a long time. Look, it's true. Elevator men and you know, there were always a man in the elevator, maybe occasionally a woman who pushed the buttons for the elevator. Are those people out of a job because of automation? Yes. Yes. But they probably needed to be out of a job. That job didn't, that was dull and tedious and it didn't need to be there. Now are we at a place where we might lose a few jobs here or there to automation? I would say yes. Particularly in factories and manufacturing. Again, these are jobs that destroy the dignity of humans.

Missy Cummings (44:57):
I would love to get the people who are having to pack my Amazon boxes. This is really, really boring work for them. It causes repetitive distress injury. I would love to get that job automated as quickly as possible. It turns out it's very hard to automate. The human hand is a thing of genius. Eventually we will start to see more and more jobs automated as we figure it out. But every time we automate one job, it opens the door to 10 more. And I think that that's what people don't realize is that we can't have enough manufacturing workers right now. We are in such a glut of labor workforce; pilots, like, you wanna be a pilot? Go sign up because we don't have enough pilots right now! So I think that people tend to hear the worst when they hear about robots are coming. I will tell you, taxi drivers, you do not have anything to worry about. Truck drivers. You do not have anything to worry about. Like

President Gregory Washington (45:51):
So you got a long time before the vehicles start driving themselves and take your jobs.

Missy Cummings (45:56):
Oh, we are so far away from that. And what's happening is we are seeing the creation of so many more jobs. And I'll tell you something else that the audience, if you're looking for a good stock tip, start your own robot maintenance company because we can't keep them all working. For the manufacturing robots that are out there, a lot of them have to sit into a closet because they don't have enough people to come and fix them when they inevitably break down.

President Gregory Washington (46:21):
Outstanding. Well this has been fantastic and I cannot wait to see the results of your research. And that will do it for this episode of Access to Excellence. I'd like to thank my guests, Professor Missy Cummings, who directs �鶹��Ƶ's Center for Robotics, Autonomous Systems, and Translational AI for taking the time to speak with me. I am �鶹��Ƶ President Gregory Washington saying, until next time, be safe �鶹��Ƶ Nation.

Missy Cummings (46:54):
And don't speed in your autopilot.

President Gregory Washington (46:56):
Alright. Of course.

Outro (46:58):
If you like what you heard on this podcast, go to podcast.gmu.edu for more of Gregory Washington's conversations with the thought leaders, experts, and educators who take on the grand challenges facing our students graduates in higher education. That's podcast.gmu.edu.

In This Story

Gregory Washington

Missy Cummings

Access to Excellence Podcast Episodes

Podcast: A small cup with big impact in the fight against Lyme disease
July 14, 2025
Podcast: A Future We Want: One Grand Challenge. Six Grand Solutions
June 3, 2025
Podcast: Building community and conversation through the arts
April 21, 2025
Podcast: Peace building amid the rise of global conflict
March 17, 2025
Podcast: James Baldwin’s insights on American life and identity
February 17, 2025

Topics

Podcast Episode

podcast

�鶹��Ƶ

Artificial Intelligence

College of Engineering and Computing

Volgenau School of Engineering

School of Computing

Center for Robotics, Autonomous Systems, and Translational AI