Angela Schoellig



Canada Research Chair in Machine Learning for Robotics and Control

Tier 2 - 2018-01-01
University of Toronto
Natural Sciences and Engineering Research Council

416-667-7518
schoellig@utias.utoronto.ca

Research involves


Combining key technologies from engineering and computer science—including control theory, machine learning and optimization—to develop the next generation of safe, high-performing robots.

Research relevance


This research will develop software that can enable service robots to operate in real-world situations in transportation, energy production, environmental sectors and more.

Programming the Next Generation of Robots


Robotics is the world’s fastest growing industry: as demand rises, the next generation of robots promises self-driving cars, drones and personal robots that can operate alongside us in unpredictable and changing environments. But to make that future a reality, we need better guarantees that these robots can operate safely and effectively in complicated real-world situations.

Dr. Angela Schoellig, Canada Research Chair in Machine Learning for Robotics and Control, is addressing this challenge. She is known globally for her work in programming high-speed, high-accuracy robots for self-driving and aircraft control. Her specialty is adapting the concept of machine learning to robotic control—that is, computer science techniques that use large amounts of data to enable computers to make step-by-step improvements on specific tasks on their own. Called control algorithms, these advanced techniques give the robot information about its changing environment, ensuring it adapts to those changes safely during operation.

Errors in robot decision-making can lead to unexpected behaviour, poor performance and instability. Schoellig and her research team are developing algorithms and software that eliminate even the smallest of errors. Their aim is to enable Canadian companies to build next-generation service and personal robots with better abilities and safer operations—and in less time with lower costs.

Based on extensive experimental testing and offering effective technology transfer, Schoellig’s approach is rooted in real-world applications that support people, companies and society. Helping to solve challenges in transportation, energy production, sustainability and the environment, her research has direct impacts on how we work and live alongside technology.