Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Launch Supplement

Queen's University

$12,500.00

Responsive Autonomy for Robots in Complex Moving Environments

EF 2024-25

Résumé
Autonomous robots will revolutionize the modern world. Basic tasks in logistics, sanitation, etc. have already been automated and focus is shifting to more challenging applications. Canadian research on these areas will benefit Canada through both advanced technology and highly trained HQP. A common, unsolved problem in robotics is autonomy in complex moving environments. This DG will develop responsive autonomous systems for these dynamic environments by studying (i) real-time motion planning, (ii) multimotion estimation, and (iii) event-based vision. These advances will create new autonomy paradigms so that robots can safely detect and react to changes and motions in the world. This is needed for self-driving cars, autonomous surface vessels, robotic household assistants, etc. Decision making and motion planning in complex environments is too slow. As a result, robots try to build detailed maps and only plan once but this fail when the world, or their understanding of it, changes. They should instead be constantly planning from their current knowledge. This responsive planning paradigm requires motion planning algorithms that reliably find high-quality paths in real time. This DG will develop such real-time motion planning algorithms using our previous success with informed, anytime, asymptotically optimal motion planning. It will advance our understanding of planning problems and find new information and better anytime approximations for real-time algorithms. Understanding dynamic environments requires measuring moving objects. Many existing systems use tracking-by-detection paradigms to find objects that are expected to move and then measure if they have. This fails when unusual objects unexpectedly move and is not portable to different environments. This DG will instead develop real-time multimotion estimation algorithms that prioritize motion. It will build on our foundational Multimotion Visual Odometry (MVO) work to improve measurement accuracy and achieve real-time performance by developing new multimotion-specific formulations and technologies. Understanding dynamic environments also require appropriate sensing. Frame-based cameras struggle to measure fast motions due to the limitations of discrete frames. Event-based cameras achieve high effective frame rates by only detecting visual change but the resulting asynchronous data is incompatible with traditional computer vision. This DG will advance core event-based vision techniques, such as event detection and tracking, for pipelines like our event-based visual odometry. This DG will fund timely and novel work at the edge of robotics to the advantage of Canadian agriculture, industry, national defence, etc. It will create portable autonomy techniques that address problems of motion and seek out industrial and governmental partners to test them on robots in the real world. It will also train highly skilled HQP in Canada based on my previous experience doing so at Oxford.
Programme
Discovery Launch Supplement
Bénéficiaire
Emplacement
Ontario, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
DGECR-2024-00017
Name nom
Gammell, Jonathan
Department département
Department of Electrical and Computer Engineering
Organization id
29
Institution établissement
Queen's University
Province en
Ontario
Province fr
Ontario
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
12500
Program id
DGECR
Program name en
Discovery Launch Supplement
Program name fr
Supplément Tremplin vers la découverte
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1507
Committee name en
Computer Science
Committee name fr
Sciences informatiques
Area of application code
700
Area of application group en
Transportation systems and services
Area of application group fr
Services et systèmes de transport
Area of application en
Transportation systems and services
Area of application fr
Services et systèmes de transport
Research subject code
2600
Research subject group en
Robotics
Research subject group fr
Robotique
Research subject en
Robotics
Research subject fr
Robotique
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Responsive Autonomy for Robots in Complex Moving Environments
Keywords
robotics; mobile robotics; autonomous robotics; autonomy; motion planning; multimotion estimation; sampling-based motion planning; almost-surely asymptotically optimal motion planning; informed motion planning; event-based vision
Application summary
Autonomous robots will revolutionize the modern world. Basic tasks in logistics, sanitation, etc. have already been automated and focus is shifting to more challenging applications. Canadian research on these areas will benefit Canada through both advanced technology and highly trained HQP. A common, unsolved problem in robotics is autonomy in complex moving environments. This DG will develop responsive autonomous systems for these dynamic environments by studying (i) real-time motion planning, (ii) multimotion estimation, and (iii) event-based vision. These advances will create new autonomy paradigms so that robots can safely detect and react to changes and motions in the world. This is needed for self-driving cars, autonomous surface vessels, robotic household assistants, etc. Decision making and motion planning in complex environments is too slow. As a result, robots try to build detailed maps and only plan once but this fail when the world, or their understanding of it, changes. They should instead be constantly planning from their current knowledge. This responsive planning paradigm requires motion planning algorithms that reliably find high-quality paths in real time. This DG will develop such real-time motion planning algorithms using our previous success with informed, anytime, asymptotically optimal motion planning. It will advance our understanding of planning problems and find new information and better anytime approximations for real-time algorithms. Understanding dynamic environments requires measuring moving objects. Many existing systems use tracking-by-detection paradigms to find objects that are expected to move and then measure if they have. This fails when unusual objects unexpectedly move and is not portable to different environments. This DG will instead develop real-time multimotion estimation algorithms that prioritize motion. It will build on our foundational Multimotion Visual Odometry (MVO) work to improve measurement accuracy and achieve real-time performance by developing new multimotion-specific formulations and technologies. Understanding dynamic environments also require appropriate sensing. Frame-based cameras struggle to measure fast motions due to the limitations of discrete frames. Event-based cameras achieve high effective frame rates by only detecting visual change but the resulting asynchronous data is incompatible with traditional computer vision. This DG will advance core event-based vision techniques, such as event detection and tracking, for pipelines like our event-based visual odometry. This DG will fund timely and novel work at the edge of robotics to the advantage of Canadian agriculture, industry, national defence, etc. It will create portable autonomy techniques that address problems of motion and seek out industrial and governmental partners to test them on robots in the real world. It will also train highly skilled HQP in Canada based on my previous experience doing so at Oxford.