Robotic manipulation: Planning sequences of pick, place and push operations

When a part sits awkwardly at the bottom of a bin, no single grasp will let the robot place it correctly: the only way to reach it leaves the robot holding the part by the very face that should end up underneath. The way out is to set it down, push it, and pick it up again, that is, to break an impossible task into a sequence of simple ones. Finding that sequence efficiently is an open problem in robotic manipulation, since the search space grows exponentially with the number of steps. This project develops planning methods that produce such sequences for previously unseen objects, in varied environments, using standard industrial grippers.

These methods serve bin picking, reorienting parts before insertion, sorting in recycling facilities, laboratory sample handling, and any task where the object arrives in a pose dictated by chance rather than by a calibrated feeder.

The testbed combines a Franka Research 3 arm, a depth camera, bins of loose parts and an instrumented work surface, backed by MuJoCo simulation to explore millions of candidate sequences before executing them.

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.