- November 15, 2021
“The main goal of my work was to develop methods and create a testing framework to evaluate the quality of Simultaneous Localization and Mapping (SLAM) algorithms,” Molodan says, “which are used to determine where you are located in an environment while also mapping that environment.”
Molodan spent the first half of her internship setting up an environment that could run the algorithms and display the results they produced (maps and calculated error) at the click of a button, and then went on to implement a method to decide which algorithm produced the best maps for a given dataset. Molodan also developed an algorithm that would count the number of enclosed areas in a map. “To understand how this is useful,” she explains, “imagine that the room the robot is mapping is just a square with no obstacles except the walls. As a robot moves around, the algorithm may not detect that the robot is still scanning the same wall, so the final map may look like a series of overlaid rotated squares. In that case, one would see many enclosed areas in the shape of triangle.”
Molodan felt that working at Jetbrains was similar to working for a Silicon Valley tech company. “I learned that a 40 hour work week feels like vacation!” she shares. “I was surprised by how much I was able to accomplish at work, and I found the material (robotics and computer vision) interesting enough to explore it further at MIT.”