Supporting Spot at The AI Institute

Katie Hughes and Spot

The AI Institute is a research-driven organization with a lofty mission to solve the most important and fundamental challenges in robotics and artificial intelligence (AI).

Katie Hughes (MSR '23) is helping create those solutions.

Hughes is a software robotics development engineer at The AI Institute, where she's worked since graduating from Northwestern Engineering's Master of Science in Robotics (MSR) program. The research center houses different research teams trying to advance robot capabilities. Some teams build robotic platforms from scratch while others use existing platforms.

In her role, Hughes supports the research teams that are using Spot, the agile mobile robot developed by Boston Dynamics.

Boston Dynamics and The AI Institute were both founded by roboticist Marc Raibert.

"We have a large stack of ROS (Robot Operating System) 2 code that many researchers use as the foundation of their project work," Hughes said. "I am one of the maintainers of this code, and I work to fix bugs and add features. I also help maintain an open source repository we use as a basis for operating Spot with ROS."

One of Hughes' recent projects involved low-level API access to Spot that Boston Dynamics recently released. Previously, users could only command the robot with high-level commands, such as "walk to this location" or "move your arm to this point."

The new release allows users to control each of the robot's joints individually.

"I helped bring this functionality to ROS 2 through contributions to our open source repository," Hughes said. "This feature is now being used by some of our researchers in their projects, who want more customization in how they control the robot. It is also available to the public."

Hughes, who interned at Boston Dynamics during her time in MSR, was excited to join The AI Institute because her responsibilities mirror those she had during her final project at Northwestern. In that case, Hughes navigated a quadruped Unitree Go1 robot — a four-legged robot that looks a bit like a mechanical dog — through a large crowd of people without colliding with anyone or anything.

"I work with researchers to implement things on hardware and with ROS 2, which is at a high level almost exactly what I did in my final project," she said. "MSR's emphasis on learning ROS 2 has really helped me in my current role as it's something I use every day. I also got a lot of technical experience on how to structure a complex project."

Hughes recently showed off that ability to create structure amid complexity by partnering with former classmate David Dorf (MSR '23) to win the NASA Space ROS Sim Summer Sprint Challenge.

Space ROS is an open-source software framework that was created to be compatible with the

demands of space robotics applications. Entrants were asked to help make Space ROS

more robust in order to help with the software's usability.

The project, much like her work at The AI Institute, required Hughes to rely on lessons learned during her time in MSR.

"My final project taught me a lot about setting checkpoints for myself and how to communicate effectively when I am stuck on something," she said. "The project-based curriculum and the experience of getting things to work on a physical robot are important skills that are fundamental to my job currently."