Certifiable Robot Perception: from Global Optimization to Safer Robots

July 12, 10AM-1PM EDT


Overview

Notice: Due to COVID-19, this tutorial will be virtual and accessible via the RSS portal (Session WS1-3).

This tutorial gives an in-depth introduction to global optimization tools, including convex and semidefinite relaxations, applied to robot perception problems. The first goal of the tutorial is to motivate the need for global solvers by providing real-world examples where the lack of robustness results from the difficulty in solving large optimization problems to optimality. The second goal is to provide the attendees with basic mathematical and algorithmic concepts, and survey important recent advances in the area. The third goal is to outline several open research avenues: global optimization has an enormous untapped potential and it is hoped that this tutorial will inspire researchers to use modern optimization tools to solve several outstanding challenges in geometric robot perception. This aims to replicate the success of the “twin” tutorial “Global Optimization for Geometric Understanding with Provable Guarantees” (held at ICCV’19), which attracted more than 150 attendees.


Organizers

Luca Carlone

Luca Carlone

Assistant Professor

Massachusetts Institute of Technology

Tat-Jun Chin

Tat-Jun Chin

Associate Professor

The University of Adelaide

Anders Eriksson

Anders Eriksson

Associate Professor

University of Queensland

Heng Yang

Heng Yang

PhD Candidate

Massachusetts Institute of Technology


Schedule

Time Topic Speaker
10:00-10:10 Arrival and Welcome  
10:10-10:45 Convex Relaxations and Strong Duality: Theory and Practice Anders Eriksson
10:45-10:55 Break  
10:55-11:30 Convex Relaxations for Certifiable Perception: Fast and Exact Global Optimality Luca Carlone
11:30-11:40 Break  
11:40-12:15 Outlier-robust Geometric Understanding: Algorithms and Provable Guarantees Tat-Jun Chin
12:15-12:25 Break  
12:25-13:00 Certifiably Robust Geometric Perception with Outliers Heng Yang