SciTech

15th annual Mobot races entertain all

Onlookers watch on as a Mobot races through one of the gates during the 15th annual Mobot races last Friday. Onlookers watch on as a Mobot races through one of the gates during the 15th annual Mobot races last Friday.

We’ve all seen it — the white line snaking its way down the steep ramps past Doherty and Wean Halls. However, not many students know that it’s actually there for a reason, unlike the snowman statue outside the aforementioned building. The white painted line is an integral part of Mobot races, a less popular though still important Carnival tradition that has been going on for 15 years now.

Mobot races center on an autonomous, student–or–staff built (depending on the division) mobile robot, which must follow the white line down the hills and through the gates. Many robots fail to make it through the first few gates; however, this year, some Mobots did make it through the course, or at least fairly far through, and were awarded prizes.

In the undergraduate division, the first, second, and third place teams were awarded prizes, and in the open division, the award went to the top team. The undergraduate awards were given to Team Zoidberg, Team REWIND, and Team 108, respectively, and the open-class award went to Team Mike and Jeff, who also set a new course record, breaking their own previous record from the year before. The first-place winner received $1000, the second $500, and the third $250, with the open-class team receiving $250 in addition to another $250 prize for setting a course record.

All Mobot winners, in addition to the cash prizes, have their names inscribed on a Mobot plaque in the Engineering and Science Library. The Mobots’ designs, while all different, are all similar in their use of some type of line-detection algorithm. This allows the Mobot to go through the course by using the algorithm to sense whether the Mobot is able to “see” the painted line, and thus whether it is going on the right path. And this algorithm is made more complicated by the fact that, at one point in the course, the painted line splits into two separate curves, which together form figure eights, and the Mobot must decide which path to take and keep following the correct path to the finish line.

First-place team Zoidberg consisted of Mark Hahnenberg and Jake Poznanski, both sophomore computer science majors. Poznanski explained that the key to their design was the mobot’s ability to be easily taken apart and put back together. This year was the team’s first year at the Mobot races, and they plan to come back next year with some improvements to the design. “We didn’t get to make the decision-making part of the algorithm working, so we’d like to fix that for next year,” Poznanski said.

Members of Team REWIND were Daniel Benjamin, a sophomore electrical and computer engineering major; Wil Hamilton, a sophomore mechanical engineering major; Clarissa Hann, a first-year mechanical engineering major; and Richard Hong, a junior computer science major. Hamilton explained that his team’s design went through changes throughout the design process. “We started out with a totally different design, and changed it a little bit at a time. We also had to completely change the motors.” Hamilton also stated that the motors were something that his team would like to change for next year’s race, to make the mobot go faster. Team 108 had teammates Rentaro Matsukata, a first-year electrical and computer engineering major, and David Soamchand, a first-year business administration major.

Matsukata explained that the team’s mobot had drive wheels in the back, and front wheels that rotated with the motor, along with two sensors to allow the Mobot to stay on the line. The team wrote the software as they went along, and wasn’t even sure it would work because they didn’t get to test it at all, finishing only about 15 minutes before the races began. In the open class, the winning team was made up of Robotics Institute staff members Michael Licitra and Jeff McMahill. Their design used four-wheel steering, 16 infrared sensors, a suspension system, and a filtering system that helped the mobot detect one or two of the white lines and pick the center, left, or white line. “Our design was based on an RC car, which is fairly fast and gives you lots of parts for pretty cheap. It’s also easy to stack components on top of the model,” McMahill said. The team would like to compete in future Mobot races and increase the speed of their mobot again, though they’re not sure how to do so at this time.