NUWC Division Newport Innovators Use Spot in Machine Learning Challenge > Naval Sea Systems Command > Saved News Module
Six teams of engineers and scientists from the Naval Undersea Warfare Center Division Newport gathered at a warehouse at Naval Base Newport for the “Spot Robot Warehouse Challenge Innovation Event” culminating the competition on June 21. The month-long challenge provided an opportunity to demonstrate an effective end-to-end machine learning pipeline.
The process allowed teams to develop technology solutions to some fundamental issues such as item identification and tracking, detecting misplaced items, and identifying missing or altered items in a warehouse environment. The teams, assisted by photographers from the Newport division, gathered images, did pre-process planning with the appropriate tools, tagged hundreds of images, built/implemented the right machine learning model architecture, verified performance, then deployed this solution “in production” on Spot. , a robot dog.
A product of Boston Dynamics, Spot is a four-legged robot designed to automate routine inspection tasks and data capture safely, accurately, and frequently. Engineers Gary Huntress and Eugene Chabot of the Platform and Payload Integration Department of Undersea Warfare (USW) acquired Spot so that scientists and engineers in the Newport division could develop their skills in autonomy. For this challenge, the teams used Spot to identify objects of different sizes, shapes, and visual characteristics.
The U.S. Navy faces many complex and dynamic environments, from detecting mines in congested landscapes to identifying changes to our critical infrastructure, event organizers noted. Navy logistics is an example of a critical function to provide the materials needed to keep the fleet able to fulfill its mission.
“The NAVSEA Logistic Command warehouse located at Naval Station Newport is a complex and dynamic environment requiring a lot of coordination to meet this challenge,” Huntress said. “Autonomy and artificial intelligence offer new opportunities to reduce labor demands and potential human errors.”
The event successfully demonstrated the teams’ ability to work on algorithms that are the basis of artificial intelligence and machine learning. Support from the Philadelphia Division of the Naval Surface Warfare Center (NSWC PD), the University of Massachusetts Lowell, and Brown University—all of whom brought their own Spot robots to take on the challenge—provided collaboration at the beyond the Newport Division science and technology community. At some point, the War Center teams will collaborate to explore the idea of leveraging each other’s strengths. The NSWC PD would use its advanced autonomy as a sensing platform while the Newport division would provide a machine learning classifier via a docker container suitable for deployment on its Spot.
Adam Sherman of the USW Platforms and Payload Integration Department, hosted the event as part of his three-month NAVSEA Journey-Level Leadership (JLL) rotation.
“When I was approached to host a ‘Spot Robot Warehouse Challenge Innovation Event’ as part of my JLL rotation, I had no idea what I was getting into, but it seemed interesting and different,” said said Sherman. “I could never have imagined the amount of work and detail that would go into this effort to make this event a success. I was lucky to have a dedicated and always available team. Each member of the team had different attributes that were monumental to ensure that whatever needed to be accomplished was completed.
Each participant was expected to have skills in Python programming, imaging, and autonomy-based machine learning. Each participant was also required to commit 32 hours to the event, which included the training, the help sessions, and the challenge event. Teams had to work within the time allotted for the challenge and had limited access to the warehouse.
The heart of the challenge was based on identifying objects in different ways. There was a fixed set of exactly 30 known objects and a well-defined autonomy task. Teams were told what they had to submit after each race and they were scored as evenly as possible. Some tasks were more difficult than others and more points were awarded accordingly. Because the Innovation Event was also a competition, the event organizers designed a challenge and scoring grid.
“We challenged 30 people to take a robot they had never used into an unfamiliar environment and teach it to search for and identify objects using a set of machine learning techniques. that most had never tried,” Huntress said. “The easiest approach would have been to use Spot’s built-in autonomy to record a predefined path and use the basic pre-trained (but not optimized) machine learning model provided to them. None team did this. Every team worked hard to extend autonomy and build a better machine learning classifier and every team succeeded.
“It’s important because the pipeline is then very easy to modify,” Huntress said. “If a new object of interest were identified, then it would be easy to get/label new images and re-run the model training without modification. In other words, the pipeline is more important than the machine learning itself. same. “
Ultimately, the fact that the teams had a better understanding of the end-to-end machine learning pipeline made the event worthwhile. As possible transitioning to US Navy use, this event showed that Spot could assist with shipyard maintenance duties in addition to warehousing.
“The result was everything I hoped for. I got this ‘aha!’ when I saw how much more valuable that pipeline was,” Huntress said. “There are nuances in the autonomy aspect. You can have a lot of tools in the toolbox, but as long as you don’t didn’t put them in the pipeline, you got nothing.
NUWC Newport is the nation’s oldest warfare center, tracing its heritage to the naval torpedo station established on Goat Island in Newport Harbor in 1869. Commanded by Captain Chad Hennings, NUWC Newport maintains significant detachments in West Palm Beach, Florida and Andros Island in the Bahamas, as well as testing facilities in Seneca Lake and Fisher’s Island, New York, Leesburg, Florida and Dodge Pond, Connecticut.
Join our team! NUWC Division Newport, one of Rhode Island’s 20 largest employers, employs a diverse, highly skilled, educated, and skilled workforce. We are always looking for engineers, scientists and other STEM professionals, as well as talented experts in business, finance, logistics and other support services who want to be at the forefront of underwater research and development. Please connect with NUWC Division Newport Recruiting on this site- https://www.navsea.navy.mil/Home/Warfare-Centers/NUWC-Newport/Career-Opportunities/ and follow us on LinkedIn @NUWC-Newport and on Facebook @NUWCNewport.