Estimation, Search, and Path Planning

//Estimation, Search, and Path Planning

ORI is expanding! I’m starting a new research group focused on estimation, search, and path planning. It’s a continuation of my previous work developing and deploying theoretically well-founded robotic algorithms here and at the University of Toronto.

As is the ORI way, we will be testing the work on a variety of real-world robotic applications. This will include our new aerial platform, Flighty.




The first publication from our new group is appearing this year at ICRA in Australia. It is the work of Rowan Border, a D.Phil. candidate from the AIMS CDT who joined ORI in 2016. It presents an autonomous, data-driven approach for surveying scenes that we call the Surface Edge Explorer (SEE). SEE finds the next best view (NBV) to improve a scene model by considering only the density of the previous measurements. This requires no specialized information about the environment and is effective on subjects of varying size and complexity.

SEE Results

The results of using SEE on a 1m dragon (left) and a 40m model of the Radcliffe Camera (right).

You can watch a video of it below or check out the paper “Surface Edge Explorer (SEE): Planning Next Best Views Directly from 3D Observations“. We are currently hard at work getting it ready to test on Flighty.


The group is continuing to grow with new research projects and D.Phil. students so I am always looking for motivated people who want to understand and solve challenging problems in robotics. If this sounds like you, please get in touch.