Designed by students for their senior engineering capstone, Mobile Virtual Player simulate tackling and reduce injury risks for athletes

The sun is setting on Memorial Field, filling the sky with wispy pink clouds as the stadium lights cut through the twilight. As the football players take the field, dance music blares at top volume from speakers set up on the sidelines, pumping up the players before the final game of the season—a high-stakes championship bout against Princeton.

One player stays behind, pacing restlessly back and forth on the running track, before being called in for a drill. At last, he zips onto the field and takes up a position facing a line of hefty linebackers. The whistle blows and the first player barrels down the field, hitting him with a loud crack and dropping him to the grass. In a second, he’s standing again, popping up like a Weeble and doing a theatrical little spin before facing down the next player in line.

Every year, emergency rooms treat 175,000 sports-related concussions in the United States, with the highest percentage from football. Research has shown that high school and college players suffer more concussions in practice than in games. Yet, this player seems unharmed by the hit. That’s because “he” is actually a Mobile Virtual Player (MVP), a remote-control tackling dummy created by a group of Thayer students to simulate tackling without putting a real person at risk.

The device started its life as a Bachelor of Engineering capstone project in the fall of 2013 and has since appeared on CNN and NPR, as well as being featured on the Late Show with Stephen Colbert—where the host, comically dressed in helmet and pads over his suit, performed his own tackle of the five-foot-tall dummy. Inquiries have started coming in from NFL, college, and high school teams eager to use the equipment in their own practices to cut down on player injuries. “We definitely saw the potential for this, and hoped it would take off,” says Quinn Connell ’13 Th’14, one of the students behind the project. “But I don’t think anyone was expecting how quickly it would move forward.”

Calling the plays on the field today is Dartmouth head football coach Eugene “Buddy” Teevens ’79, who has been a crusader for reducing injuries at practice—going so far as to ban full-contact practices back in 2010. Three years ago he was talking with his classmate John Currier ’79 Th’81, a Thayer research engineer, about mechanical ways the team might simulate tackling. “We decided the most effective way would be to take it to the Thayer School and position it as a capstone project,” says Currier, who reached out to Elliot Kastner ’13 Th’14 ’15, an engineering student and Dartmouth football player. He was enthusiastic about taking it on. “For the majority of my life, football and engineering have been living in separate worlds,” says Kastner, clutching the MVP’s remote control on the sidelines. “Now I got to see them come together.”

Joined by Connell and fellow students Andrew Smist ’13 Th’14 and Noah Glennon Th’14 ’15, the group blue-skied possibilities, considering all of the ways they could create something that would give players the experience of tackling while keeping them safe from injury—including drones, helicopters, and even using animals. “Should it float? Should it fly? Should it be shot out of something? We considered all of the possibilities,” says Kastner.

Finally, they settled on taking the classic football tackling dummy—but updating it to make it mobile. As they started researching, they found that the football dummy had been invented in 1936 and hadn’t been updated much since. The students interviewed players and coaches to figure out what they might be looking for, deciding on a humanoid form so players could practice proper form, leading with the shoulder rather than the helmet. They had players test different materials for tackling, zeroing in on a firm foam for the base and a softer moldable foam, surrounded by a vinyl cover, for the top.

Early on, the students decided the dummy would need to be self-righting, preferably within two seconds of being tackled, in order to maximize the number of drills in practice. That presented particular design problems. “The whole balance of the weight was a big issue,” says Smist. “We wanted to have enough weight in the bottom so it could stand up on its own and have enough weight in the top so they weren’t flying through it.”

The electronics they initially used in the drivetrain turned out to be too complex, constantly malfunctioning due to the repeated hits the MVP suffered. “It was definitely a delicate balance of trying to find that performance we wanted and still making it robust enough to be repeatedly tackled,” says Connell, adding, “How many engineering projects do people work on where the end goal is to make something that can be beat up?”

Kastner estimated that over three years of practices, the MVP would take 6,000 tackles. “My computer couldn’t take 6,000 tackles,” he says. Adding to the difficulty, the dummy had to be cheap enough that it could be mass-manufactured and affordable to college and high school teams working within limited budgets. “We could build a $10 million anatomic robot, but that wouldn’t be that useful,” says Kastner.

Currier was advising the project in a hands-off manner, letting students create their own design—and make their own mistakes. “There was one time when an early prototype was running around, and there was a fair amount of smoke coming out of it,” he laughs. “They were really persistent in coming up with something that would work.” By graduation in the spring of 2014, the team had completed a design and a number of elements but still didn’t have a fully functioning prototype. Kastner and Connell decided to continue working on the project, spending the summer of 2015 in Hanover working on the project in space at Thayer, repeatedly iterating different ideas. At one point, they were trying to engineer a new motor for the drivetrain, and unable to find anything that would work. In one of those “aha!” moments, says Connell, “we looked at the drill we were using to put things together. In a half an hour we were using the drill, and it worked.”

They ended up with a device that, for all of its bulk, is surprisingly agile. After the first tackle this afternoon, another linebacker charges ahead for his own takedown. But as he barrels down the field, the MVP skirts away from the attack, outrunning the player, who hits the turf empty-handed. The MVP does an extra spin, as if to drive home its victory.

One player who didn’t miss his tackle with the MVP was Madison Hughes ’14, who played rugby with Connell at Dartmouth before going on to captain the U.S. Olympic rugby team. Running into Connell and Kastner one day as they were testing their prototype, he made a short video of himself tackling the dummy and posted it online. Within 48 hours, the video had one million views, and suddenly the students’ phones started ringing with media who wanted to interview the inventors and coaches who wanted to buy an MVP.

“It was an amazing experience,” says Kastner of appearing on the Late Show, where he bantered with Colbert along with Teevens and watched Connell pilot the MVP across the floor of the Ed Sullivan Theater as Colbert dove in for the takedown. All the media attention, however, suddenly meant that the students couldn’t continue to tweak the device in private. “The publicity has really put the pressure on us to get it out to the market first,” says Kastner.

The four students worked on a business plan with Tuck MBA student Alex Jenny ’10 Tu’16, a former Dartmouth quarterback, and gained advice from the Dartmouth Entrepreneurial Network. Then Teevens, Kastner, and Connell formed a company, Mobile Virtual Player LLC, to take the project forward.

Asked if he thinks the MVP will make him rich, Kastner answers with the diplomacy of a winning quarterback after a game, saying, “Protecting players is the most important thing.”

In the meantime, the now-former students are refining their prototype. They showed it at the American Football Coaches Association convention in January, and they hope that beta versions of the MVP will hit the field for testing by other teams in the spring. They are also looking further ahead. Mobile Virtual Player LLC sponsored two new 89/90 teams this year to help develop the next generation of the robot. Advancements could include attachments for use in other drills—for example, a hoop that a quarterback could use to practice throwing—and more sophisticated electronics that would make the MVP self-driving. “Right now, it is remote control, but we are working on making it programmable,” says Glennon. “Eventually, a team of four or five of them could all be programmed to run a certain drill together.”

Sometime in the future, a team of MVPs may be able to play a football game by themselves. Until then, they’ll have to content themselves with keeping their flesh-and-blood teammates free from injuries.

—Michael Blanding is a Boston-based writer and the author of The Map Thief. This article originally appeared in the Winter 2016 issue of the Dartmouth Engineer. Since publication, the MVP is widely used in the NFL and college football teams nationwide.