Max Robotics — Global Service Robot Directory + US Certification

About

Connor Holmes is a leading researcher in certifiable state estimation and perception for robotics, with a focus on provably optimal solutions for challenging non-convex problems. His work has pioneered the use of semidefinite relaxations and convex optimization to guarantee global optimality in landmark-based SLAM and range-only localization—areas where traditional methods can only promise local convergence. Holmes’s most influential paper, “An Efficient Global Optimality Certificate for Landmark-Based SLAM” (2023, 22 citations), provides a practical certificate for verifying global solutions, while his “Safe and Smooth: Certified Continuous-Time Range-Only Localization” (2022, 17 citations) extends these guarantees to dynamic, continuous-time settings. He has also advanced the theory of certifiable perception with “On Semidefinite Relaxations for Matrix-Weighted State-Estimation Problems in Robotics” (2024, 9 citations) and “Toward Globally Optimal State Estimation Using Automatically Tightened Semidefinite Relaxations” (2024, 8 citations), which automate the construction of tight convex relaxations. Holmes’s contributions are foundational for safe, reliable autonomy, enabling robots to operate with mathematical certainty in uncertain environments.