Joint prediction of grasp-placement feasibility in robotic manipulation

How a robot picks up a part determines what it can do with it next: a poorly chosen grasp makes an insertion or a tidy placement impossible, because the robot's hand ends up exactly where the part needs to go. Conventional methods plan the grasp first and the placement afterwards, and therefore waste time on combinations that were doomed from the start. This project develops a learned predictor that scores the feasibility of a grasp-placement pair in one shot, across several candidate final poses, accounting for collisions and for what gravity does the moment the object is released.

Applications include kit assembly, insertion of parts into fixtures, machine tending, bin picking and packaging. In all of these, a robot that can handle parts it has never seen, without reprogramming, greatly lowers the integration cost by increasing the adaptiveness.

Experiments run on a Franka Research 3 station with a parallel-jaw gripper, a wrist-mounted depth camera and a force-torque sensor, fed from bins of machined and 3D-printed parts.

How to apply

To apply, send a complete application for the NSERC Canada Graduate Scholarship (CGS) Doctoral Program (written in English) to Prof. Philippe Nadeau. The CGS is the primary funding mechanism for PhD students in Canada, and competition for it is fierce. Preparing a full application up front may look like extra work, but it pays off: you'll arrive with a strong first draft of your CGS application well ahead of the deadline, improving your odds of being funded by the time you join the group.

Note: If you are considering applying for a Master's with thesis, the same steps apply but we will be using your application package to apply for the appropriate funding sources.