convex relaxation

A simple, flexible, and computationally-efficient approach to robust synchronization over groups.

A simple and computationally efficient approach to optimizing the design of sensor systems for robotic perception.

We develop the first convex relaxation for the general multi-robot range-aided SLAM (RA-SLAM) problem.

We develop the first convex relaxation for the general multi-robot range-aided SLAM (RA-SLAM) problem.

Convex (specifically semidefinite) relaxation provides a powerful approach to constructing robust machine perception systems, enabling the recovery of certifiably globally optimal solutions of challenging estimation problems in many practical …

An elegant and robust approach to solving challenging inverse kinematics problems using a distance-geometric problem formulation & convex relaxation.

The first *distributed* algorithm provably capable of recovering correct (*globally optimal*) solutions of SLAM and rotation averaging. Honorable Mention, IEEE Transactions on Robotics King-Sun Fu Memorial Best Paper Award

A fast algorithm for *certifiably globally optimal* rotation averaging. Implemented in the GTSAM library. ECCV 2020 spotlight talk (top 5%)

Build your own certifiably correct machine perception methods

The first practical algorithm *provably* capable of recovering correct (*globally optimal*) solutions of the SLAM problem. Invited article (IJRR Special Issue)