Weather Forecast


Duluth native part of college team that built hands-off snowplow

From left; Brianna McIntyre, Alex Beaulier, Brandon Jameson and Josef Mendez, all University of St. Thomas graduates, demonstrate their autonomous snowplow on campus in St. Paul on Wednesday. (Craig Lassig / for the News Tribune)1 / 3
Josef Mendez of Bloomington, Minn., a University of St. Thomas graduate, talks about his team's autonomous snowplow - a robot that can remove snow from a driveway - on campus in St. Paul on Wednesday. (Craig Lassig / for the News Tribune)2 / 3
Josef Mendez of Bloomington, Minn. (from left), Alex Beaulier of Duluth and Brianna McIntyre of Maplewood, Minn., all University of St. Thomas graduates, show off their autonomous snowplow on Wednesday. (Craig Lassig / for the News Tribune)3 / 3

ST. PAUL — Minnesota is affectionately referred to as "Minne-snow-ta" for a reason. Waking up to 6 inches or more of white fluff and getting into work two hours late because of said fluff is not an uncommon experience during the winter months.

Now imagine: Instead of pulling on long underwear and heavy-duty waterproof boots first thing in the morning, you sip coffee in the warm comfort of your bathrobe while a robot snowplow clears the driveway.

If technology is inescapable, at least it can do the dirty work.

That's the idea that five University of St. Thomas (UST) graduates, including one Duluth native, have brought to fruition. The team spent two semesters building an autonomous snowplow, named "The Blizzard Byter," which will compete in the seventh annual Institute of Navigation autonomous snowplow competition this weekend as part of the St. Paul Winter Carnival.

Alex Beaulier, a Denfeld High School graduate who majored in mechanical engineering, designed and built the autonomous plow with his teammates: Josef Mendez of Bloomington, Minn., electrical engineering; Brianna McIntyre of Maplewood, Minn., electrical engineering; Brandon Jameson of St. Cloud, computer engineering; and Dustin Anderson of Chester, S.D., mechanical engineering.

Each of the team members graduated in December and completed the project as part of their senior design class.

At the competition this weekend, the UST team's autonomous plow will compete in two simulations: one test simulates a sidewalk and the other simulates a driveway with fixed and moving obstacles. These obstacles could represent a parked car or a child, Jameson said — objects a smart autonomous plow would want to avoid hitting.

The competition is designed to encourage young engineers to innovate in the world of autonomous navigation, McIntyre said. The team will compete Saturday if they pass safety checks today, she added.

"We've had lots of people offer to be test subjects for us," McIntyre said.

Kundan Nepal, an associate professor in UST's School of Engineering and a sponsor of the project, said the idea for the project began in December 2015 when the school's robotics club created a similar plow. That plow was a building block for this project, he said.

Nepal said he gave the students project requirements and they began designing their plow in June. The completed plow had to be small enough to fit through doors, move no faster than two meters per second, include safety mechanisms for emergency stops, and be able to clear snow from a standard driveway.

The design also needed to be cost-effective, Nepal said.

"We went with the cheapest possible things we could find," Nepal said.

At the competition this weekend, the team will face schools with much higher budgets, Beaulier said. While UST allotted the team $2,500 for their project, other schools have budgets upwards of $25,000.

With that budget constraint in mind, the team constructed the plow with common, commercially used dimensions so it could be easily reproduced.

"There's potential for commercialization of this technology," Nepal said.

Brianna McIntyre of Maplewood, Josef Mendez, of Bloomington, Brandon Jameson of St Cloud and Alex Beaulier of DuluthBeaulier described the autonomous snowplow as "a giant Roomba for your driveway."

The plow runs like an electric car, Mendez said. It runs on a laptop placed inside the plow's wooden body and is powered by two batteries, similar to car batteries, and two high-torque motors. The team estimated the plow weighed 300 pounds.

A laser range finder sits on top of the plow's body. The finder spins, shooting beams of light. Jameson said those beams bounce off objects nearby, helping the computer system map out the plow's surroundings. This technology is called LIDAR, he said, which stands for Light Detection and Ranging.

"The robot knows where it is in relation to this grid system," Jameson said.

Mendez said this process is similar to the cruise control setting in a car. The car's computer system adjusts the car's speed accordingly. Similarly, the plow's computer system adjusts the plow's direction if it senses the vehicle has moved too far left or right of its destination.

The team tested the plow outside at the UST campus Wednesday night. The plow, driven by Jameson with a video game controller, easily maneuvered on the snow-laden concrete. It left the ground almost clean — cleaner than most manually shoveled driveways.

Beaulier said he is not sure the plow has a limit to the amount of snow it can push.

"It's the wheels that start slipping before we hit a limit," Mendez said, referring to the plow's small wheels, similar in size to wheels on a snowblower.

McIntyre seemed more confident of the Blizzard Byter's ability to push heavy amounts of snow.

"We sacrificed speed for power," McIntyre said as the plow slowly moved back inside the Facilities and Design Center at St. Thomas.

The team hit a few roadblocks throughout their design process. Beaulier said UST's engineering program covers a broad range of topics, but the autonomous plow's mechanics required him to delve deeper into some of those subjects.

Mendez said he found the budget tough to stick to, considering all of the more expensive gadgets the group found while researching.

"But then you have to find something that's a little more practical," he said.

Jameson said if the team were to continue with the project, they would work toward a self-activating autonomous snowplow. That could include a phone app to turn the plow on while its owner is still lying comfortably in bed. Or perhaps the plow's computer system would sync with a local weather service so it would know when it is snowing.

The team said they would want the plow to be as hands-off as possible.

"That's sort of our dream and our vision with this," Jameson said.