Florian Shkurti

Florian Shkurti
Monday, November 4, 2019 - 9:58am

POSITION: Assistant Professor, Tenure Stream

ACADEMIC UNIT: Mathematical & Computational Sciences



“The possibilities are amazing. There’s not a better place to have landed.”

While Florian Shkurti’s classes as an undergraduate focused on theoretical computer science and abstract mathematics, he was eager to find a way to apply this knowledge to the “real world.” A robotics competition caught his eye and, after participating, Shkurti realized the emerging field offered the possibilities he’d been seeking.

Shkurti went on to earn a master’s degree and a PhD in computer science and robotics, and joined UTM last December as one of the professors hired to create the campus’ new robotics cluster. He is also a member of U of T’s new Robotics Institute. 

“The possibilities are amazing,” he says. “The people at UTM are extremely helpful, the campus is beautiful and the department’s administrative support is aligned to our efforts. There’s not a better place to have landed.”

Shkurti’s research focuses on the algorithms and learning techniques that make it easier for robots to collaborate with humans. He is currently working with imitation learning, a subset of machine learning that will allow robots to learn easily from human demonstrations without the need for endless repetitions of the demos. The goal is to reduce a robot’s uncertainty about what to do in various situations – driving a car, for example, or exploring unseen natural environments as a scientist would.

“You need to think about how to turn a demonstration video into driving commands, which requires collecting a lot of data from humans about how they drive and why they make certain choices,” Shkurti says.

Despite having earned his undergraduate degree at U of T, Shkurti had not visited the UTM campus prior to accepting his current position. “It is a very energizing environment and set of colleagues,” Shkurti says. “It is amazing to see our vision come to life.”

Selected Publications:

  1. Florian Shkurti, Nikhil Kakodkar, Gregory Dudek. “Model-Based Probabilistic Pursuit via Inverse Reinforcement Learning”. IEEE International Conference on Robotics and Automation, 2018.
  2. Florian Shkurti, Wei-Di Chang, Peter Henderson, Jahidul Islam, Juan Camilo Gamboa Higuera, Jimmy Li, Travis Manderson, Anqi Xu, Gregory Dudek, and Junaed Sattar. “Underwater Multi-Robot Convoying using Visual Tracking by Detection,” IEEE International Conference on Intelligent Robots and Systems, 2017.
  3. Florian Shkurti, Anqi Xu, Malika Meghjani, Juan Gamboa, Yogesh Girdhar, Philippe Giguere, Bikram Dey, Jimmy Li, Arnold Kalmbach, Chris Prahacs, Katrine Turgeon, Ioannis Rekleitis, Gregory Dudek. “Multi-Domain Monitoring of Marine Environments Using a Heterogeneous Robot Team,”IEEE International Conference on Intelligent Robots and Systems, 2012

Read about other new faculty members at UTM.