LARRI Team Captures Top Honors in First Time at NSIN Hackathon

By Holly Hinson

A student team from Speed School’s Louisville Automated Robotics Research Institute (LARRI) captured first prize at the Off the Beaten Path hackathon held on February 10. Hosted by the National Security Innovation Network, (NSIN) the goal of the hackathon event is to produce projects that develop concepts or technologies for new sensors, sensing techniques, perception, and intelligence algorithms to generate actionable information for an autonomous vehicle to successfully move through a rugged terrain.

The background of this is that autonomous off-road mobility is becoming a critical requirement for various military operations, including maneuvering in rugged environments and disaster relief. Vehicles in these missions will require sensors and software to navigate unexpected obstacles and have the hardware necessary to interact with these potentially difficult elements.

The LARRI team, known as ATVision, proposed a solution that involved a combination of adversarial attack detection, convolutional neural networks, deep learning and block chain technology. University of Louisville competed with seven other top engineering schools such as UC Berkeley, Georgia Tech, MIT and Harvard in the hackathon and took top honors in the first year ever participating in the event.

Sumit Kumar Das, a research scientist at LARRI, served as the team’s faculty advisor and helped prepare ATVision for the competition. The team members included three Computer Science & Engineering (CSE) majors; Christopher Trombley, Karthik Malyala and Bhumika Gowda, and two Electrical and Computer Engineering majors (ECE).

One member of the team, Karthik Malyala, explained the division of labor that led to success for the team. “Four of us, including Sumit Kumar Das, Christopher Trombley, Bhumika Gowda and myself contributed to the overall concept and presentation, while Henry Reynolds and Christopher Tran, as our ECEs were in charge of the actual implementation of the prototype,” said Malyala.

How did the ATVision team stand out for the judges? “We had to create a five-minute pitch video to describe our solution,” said Malyala. “I think one aspect that impressed the judges was that even though I was the only representative on the team in the judging panel, I was still able to tackle a variety of questions about the other technologies that I wasn’t personally responsible for, so that made them really understand how our project covered all disciplines and aspects of the problem.”

The judging panel, which included an Army researcher, reflects the desire of the defenses to further develop the capabilities of autonomous link tracking systems for ATVs or other rugged vehicles in order to identify obstacles and communicate with other autonomous vehicles.

Malyala explained ATVision’s three-part project started first with computer vision. Secondly, a Variational Autoencoders based autonomous navigation system was used to learn the distribution of sensor data used during autonomous navigation, which can detect adversarial attacks. The third part of the solution was a block-chain based network security, which means the team analyzed and encrypted communication between very different autonomous vehicles using block chain technology.

The potential for this kind of robotics research is wide-ranging. Malyala explained that, “Computer vision may perceive aspects of the environment that the human eye cannot see necessarily, so we can send this autonomous vehicle on a search operation, and it can probably gather more information than a human can in a given period of time.”

Malyala said LARRI is giving students like those on the hackathon team an amazing opportunity to compete on bigger stages in the robotics field. “During the judging panel, I was talking to professors from UC Berkeley, and one of the teams we were competing against were engineers from there, which is one of the top universities for computer science,” he said.

“LARRI also enabled me to utilize all this knowledge that I’ve learned in class, or with hands-on activities with projects funded by the National Science Foundation,” said Malyala. “I was thrilled to be using the technology learned at LARRI on a small scale for something on a much larger scale that can really benefit the community,” he said.

Born in India, Malyala immigrated to the U.S. in 2013, and it was in high school that his interest in the world of technology was piqued. “My father had hearing damage, and wears hearing aids,” he said. “Just seeing how technology enables people to see or hear things they’ve never done before, it really inspired me.”