Grant: Canada Research Chairs
University of Toronto
$100,000.00
Collaborative Robotics
EF 2024-25
Résumé
Collaborative robots, or cobots, have the potential for tremendous positive impact in fields as diverse as manufacturing and healthcare, but significant challenges related to flexibility, adaptability, and generalizability must still be overcome. The proposed research program will directly address these challenges by creating new learning algorithms that can be successfully applied to the high-dimensional sensing and action spaces in which these robots must operate.
Recent successes in machine learning, and particularly deep learning, have shown that lower-dimensional 'latent' representations can be discovered in a supervised or semi-supervised way. Many of these latent structures take the form of manifolds: curved, non-Euclidean spaces that, locally, look 'flat.' Manifolds (and associated metrics) are important in part because they enable one to define the ideas of `which way to go' and `how far to travel' in a rigorous way, while also reducing the dimensionality of many problems, making them tractable. Geometric deep learning, which is focused on manifolds and their embeddings, has made great strides in many areas, but its application in robotics is still emerging.
The primary objective of the research program is to advance the state of the art in human-centred robotic manipulation, moving robots out of research laboratories and caged factory environments. Three main research themes will complement and extend ongoing work in the nominee's laboratory: learning on manifolds, learning uncertainty on manifolds, and discovering manifolds from data. Fundamental research within each theme will enable robots to identify and leverage underlying structure in an operating domain and within the available sensor data, facilitating adaptation and generalization. The themes will be supported by an extensive experimental campaign involving real robots; this approach ensures that the developed algorithms will be robust and applicable in real-world scenarios. Experimental evaluation will leverage several advanced robotic manipulators available in the nominee's laboratory.
The research program has the potential to transform industries where robotic manipulation is and will be essential, such as warehousing, logistics, advanced manufacturing, and healthcare. Novel research outcomes will enable Canadian firms to deploy highly capable cobots in key international markets and help to establish Canada's leadership in collaborative robotics.
Programme
Canada Research Chairs
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data
Tel que publié
Application id
CRC-2022-00365
Name nom
Kelly, Jonathan
Department département
Institute for Aerospace Studies - University of Toronto
Organization id
31
Institution établissement
University of Toronto
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2022
Award amount
100000
Program id
CRC
Program name en
Canada Research Chairs
Program name fr
Chaires de recherche du Canada
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
9998
Committee name en
Canada Research Chair Selection Committee
Committee name fr
Comité de sélection des Chaires de recherche du Canada
Area of application code
9999
Area of application group en
Not available
Area of application group fr
Non disponible
Area of application en
Not available
Area of application fr
Non disponible
Research subject code
9999
Research subject group en
Not available
Research subject group fr
Non disponible
Research subject en
Not available
Research subject fr
Non disponible
Field of research list codes
RDF1020102; RDF1020104; RDF2030103
Field of research list names en
Intelligent robotics;Machine learning;Middleware systems
Field of research list names fr
Robotique intelligente;Apprentissage machine;Systèmes intergiciels
Socioeconomic objective list codes
RDS10610
Socioeconomic objective list names en
Information systems, technologies and services
Socioeconomic objective list names fr
Systèmes, technologies et services d'information
Application title
Collaborative Robotics
Keywords
Collaborative Robotics; Robotic Manipulation; Motion Planning; Machine Learning; Differential Geometry; Robot Perception; Deep Learning; Kinematics & Dynamics
Application summary
Collaborative robots, or cobots, have the potential for tremendous positive impact in fields as diverse as manufacturing and healthcare, but significant challenges related to flexibility, adaptability, and generalizability must still be overcome. The proposed research program will directly address these challenges by creating new learning algorithms that can be successfully applied to the high-dimensional sensing and action spaces in which these robots must operate.
Recent successes in machine learning, and particularly deep learning, have shown that lower-dimensional 'latent' representations can be discovered in a supervised or semi-supervised way. Many of these latent structures take the form of manifolds: curved, non-Euclidean spaces that, locally, look 'flat.' Manifolds (and associated metrics) are important in part because they enable one to define the ideas of `which way to go' and `how far to travel' in a rigorous way, while also reducing the dimensionality of many problems, making them tractable. Geometric deep learning, which is focused on manifolds and their embeddings, has made great strides in many areas, but its application in robotics is still emerging.
The primary objective of the research program is to advance the state of the art in human-centred robotic manipulation, moving robots out of research laboratories and caged factory environments. Three main research themes will complement and extend ongoing work in the nominee's laboratory: learning on manifolds, learning uncertainty on manifolds, and discovering manifolds from data. Fundamental research within each theme will enable robots to identify and leverage underlying structure in an operating domain and within the available sensor data, facilitating adaptation and generalization. The themes will be supported by an extensive experimental campaign involving real robots; this approach ensures that the developed algorithms will be robust and applicable in real-world scenarios. Experimental evaluation will leverage several advanced robotic manipulators available in the nominee's laboratory.
The research program has the potential to transform industries where robotic manipulation is and will be essential, such as warehousing, logistics, advanced manufacturing, and healthcare. Novel research outcomes will enable Canadian firms to deploy highly capable cobots in key international markets and help to establish Canada's leadership in collaborative robotics.