Hit enter to search or ESC to close
Robotics

Why the New Director of the CMU Robotics Institute Thinks a Self-Driving Sea Change Is Underway

Matt Johnson-Roberson of the CMU Robotics Institute. Photo Credit: Carnegie Mellon University

Photo Credit: Carnegie Mellon University

“Ironically, I don’t really drive,” admits the man who has spent his entire career working on self-driving vehicles. “I ride my bike everywhere.” 

Matt Johnson-Roberson’s little secret may at first seem counterintuitive. But allow him to explain: “I often look at how I move around the city on a bike and think, ‘Man, this is a really good way of thinking about robot problems.’”

Navigating his city by bike, Johnson-Roberson has closely observed the rigorous structure of the street as well as the rulelessness of the sidewalk. It is a system that offers a keen lesson on the flow and function of daily life—or, sometimes, the lack thereof. The experience often gets him thinking about how we can build autonomous vehicles that are not only capable of moving humans and goods fluidly between street and sidewalk, but across all facets of society. 

The last couple of years seem to have only amplified these questions. Which makes Johnson-Roberson’s new job all the more exciting. He recently took over as Director of Carnegie Mellon University’s Robotics Institute, working for a school that is also home to Argo AI’s Center for Autonomous Vehicle Research. 

“We are at this new moment where there’s just a tidal wave of new applications for robotics,” says Johnson-Roberson. “And why I think it’s so important, as an educational institution, to think about that is because students are hungry for new and exciting problems.”

Self-driving technology, Johnson-Roberson continues, is particularly significant to this new generation of roboticists—especially with focused research hubs like the CMU Argo AI Center for Autonomous Vehicle Research 

Established just two-and-a-half years ago, the Center brought together an industry leader within the burgeoning realm of self-driving and academia like never before. Through the Center, CMU students have access to Argo scientists’ data and tools, and vice versa—a partnership that has already yielded safer reinforcement learning models, more accurate and efficient vehicle localization, and lidar-based technology that drastically improves spatial understanding and prediction. 

“What excites me is that there’s now a critical mass of people at places like Argo who are working on the same problem,” Johnson-Roberson says. “That’s an incredible way of thinking about distilling all of this existing technology and ideas, and combining them in new and novel ways to go from a world in which they do not work to a world in which they do work.”

A ‘technological adventurer’ from the start

The job is a homecoming of sorts for Johnson-Roberson, who is a graduate of CMU’s School of Computer Science—the place where his love of autonomous technology took root. It’s also the perfect fit for Johnson-Roberson’s infectious enthusiasm for robotics and its potential for improving everyday life—a trait that inspired his mentor at CMU, Professor Red Whittaker, to dub him a “technological adventurer.” 

Growing up in Brooklyn, New York, in the 1980s and 90s, Johnson-Roberson always knew he wanted to work with computers because he liked video games. After his parents got him an Apple IIGS, he got hooked on programming. When he started looking around at colleges that offered computer science degrees, he discovered CMU and its Robotics Institute. Established in 1979, the Institute was the first department of its kind among American universities. “CMU had the best program in the world,” Johnson-Roberson says. “So I was, like, ‘Great, I’m gonna go there.’”

Just as he was having reservations about his concentration, video game programming, Johnson-Roberson noticed an ad for a class called the DARPA Grand Challenge, taught by Whittaker. The Challenge, funded by the U.S. government’s Defense Advanced Research Projects Agency, would be the world’s first long-distance race for autonomous vehicles. The hope was that, by utilizing the brightest minds in both academia and the private sector, the Challenge would spur the development of AV technology. 

The race was first held in March of 2004 in California’s Mojave Desert. Whittaker’s crew—Team Red—had kitted out a 1986 Hummer that ended up traveling the furthest distance among all the competition, despite not finishing the course. “It was such a ragtag bunch of people with different backgrounds who didn’t really know what was going to be involved in solving the Challenge,” says Johnson-Roberson. “It was an incredible experience that changed the trajectory of the rest of my career.” 

Finding a footing in outer limits of AV tech

After graduation from CMU, Johnson-Roberson was thoroughly infected with the adventure bug. He set out for Australia, even though he didn’t know a soul there, to get a doctorate in robotics from the University of Sydney. It was in Australia that Johnson-Roberson had a chance to reach further toward the outer limits of autonomous tech. From the Great Barrier Reef to the lesser-known, but no less spectacular, Ningaloo Reef, Johnson-Roberson experimented with deepwater robots to explore the reefs’ incredibly diverse and vital coral ecosystems. 

“I got to visit 20 or 30 countries, deploying robots in all kinds of interesting and extreme environments,” he says. “Teaching and researching in these different countries and places just expanded my view of the world.”

In 2013, Johnson-Roberson returned to the U.S. to share what he learned with students at the University of Michigan, where he founded and taught in the Deep Robot Optical Perception (DROP) Lab, which aims to develop autonomous vehicles capable of operating in the kind of dynamic environments he’d grown comfortable working in during his time at the University of Sydney. Three years later, in 2016, Johnson-Roberson was made co-director of the UM Ford Center for Autonomous Vehicles.

By 2019, it was becoming clear to Johnson-Roberson that “a sea change” was underway in last-mile delivery—the movement of goods from retailer to consumer—“particularly for what could be delivered and on what timeframes and at what cost.” So, along with friend and fellow professor Ram Vasudevan, Johnson-Roberson founded Refraction AI, a company that builds autonomous vehicles (AVs) to improve the efficiency and safety of last-mile delivery services.

Since the pandemic, the last-mile, along with every other link in the supply chain, has undergone even more radical change. This tectonic shift has made companies like Refraction AI, together with research hubs like the CMU Argo AI Center for Autonomous Vehicle Research, all the more vital to the future of how humans, and the goods we rely on, move.  

“So much amazing work has been done in robotics over the last 25, 30 years,” Johnson-Roberson says, reflecting on his career since the DARPA days. “There’s this incredible opportunity now, in seeing the coalescence of all of that work into actual deployed mobile robotic systems, particularly those that are out there interacting with humans, cars, factories, assembly lines—all these things that were outside the reach of industrial robots.”

Familiar ground, fertile future

Through the mentorship of Red Whittaker at CMU, Johnson-Roberson was able to find his place as a technological adventurer. Now, it’s his turn to mold the current moment into the shape he believes will be most transformative for society in the future. “I think that there is a big difference in this current generation of students, in that they really want to see the impact and effect of their work in the world,” Johnson-Roberson says. 

Places like the Robotics Institute and the Argo AI Center for Autonomous Vehicle Research, he believes, are perfect “jumping off points for students, scientists, and professors to smash together like electrons, to bounce off each other, to have new and weird and crazy ideas, and then try to go build them.” 

Especially today, with a crop of students that is more diverse than ever before. “We have students coming from all over the world, from a bunch of different economic backgrounds, social backgrounds, and experiences of the world,” he continues. “Why I think that’s important, and why I think that’s going to be an incredible opportunity for not just us at RI but also across the industry, is that means they have had a different set of problems, a different set of experiences, that they can bring to bear to say, ‘Hey, this is the thing that we need to solve, because it’s gonna make a huge difference.’”

For Johnson-Roberson, this all equates to stepping off into the unknown—an adventure he’s prepared for his entire life. 

Choose your lane

How Autonomous Vehicles Distinguish Between Bicycles and People Who Ride Them
Self-Driving

How Autonomous Vehicles Distinguish Between Bikes and People

When it comes to how autonomous vehicles see the world, humans come first, literally. Autonomous vehicles (AVs), like the kind operated by Pittsburgh-based Argo AI, use Machine Learning to detect and classify the objects in their surroundings, identifying people...
Why The League of American Bicyclists is optimistic about autonomous vehicles
Self-Driving

Why a Leading Cycling Advocacy Group Is Optimistic About Autonomous Vehicles

As autonomous vehicle use grows, AV companies and the League of American Bicyclists are collaborating on how to ensure cyclists and motorists can share the roads safely, even if the “motorist” is artificial intelligence software. As part of the...
Opinion

Self-Driving Is Arriving Right On Time. Just Like Ice Cream Did

Seven years ago, I was a self-driving skeptic. Not of the technology. Of all the “experts” promising autonomous vehicles would be everywhere by 2020. You didn’t need to be Nostradamus to know that was ridiculous. All you needed was...
Illustration of a futuristic parking deck turned into a mixed-use space, with AVs driving by
Business

How Autonomous Vehicles Could Help Transform Parking Lots

Researchers say it’s likely that autonomous vehicles (AVs) can help reduce the need for parking lots, opening more room for grass and trees and other elements of nature. It may not seem like it when you’re circling the block...
An illustration of an Argo autonomous vehicle in teal against a backdrop of buildings, a bicyclist, and research papers
Self-Driving

7 Big Breakthroughs From Argo Research at CVPR 2022

The 2022 Conference on Computer Vision and Pattern Recognition (CVPR 2022) is nearly here. Thousands of computer scientists, software engineers, and researchers from around the globe will gather in New Orleans to review and discuss their latest work in...
Self-Driving

Researchers Predict the Future With Lidar Data

Researchers at Carnegie Mellon University’s Argo AI Center for Autonomous Vehicle Research, a private-public partnership funded by Argo for advancing the autonomous-vehicle (AV) field, say they have come up with a way to use lidar data to visualize not...

Must Reads