Max Robotics — Global Service Robot Directory + US Certification

About

Lukas Luft is a robotics researcher specializing in probabilistic state estimation, multi-robot systems, and autonomous vehicle localization. His most influential contributions lie in the domain of decentralized collaborative localization, where he developed fully decentralized algorithms based on the extended Kalman filter that enable multiple robots to jointly estimate their poses without relying on a central coordinator. His 2018 paper addressing asynchronous pairwise communication — a particularly thorny challenge in real-world deployments — has garnered 90 citations, while his earlier 2016 work on sparsely communicating robots has accumulated 44 citations, reflecting sustained community interest in this problem. Beyond multi-robot coordination, Luft has made notable contributions to lidar-based perception, including an analytical lidar sensor model grounded in ray path information (25 citations) and a compact, differentiable map representation using the discrete cosine transform. His 2020 work on long-term urban vehicle localization using pole landmarks from 3D lidar scans demonstrates a practical orientation toward autonomous driving applications. He has also ventured into shared autonomy, offering a principled Bayesian framework that extends classical filter derivations to settings where robot controls are purposefully chosen. Across his body of work, Luft consistently bridges rigorous probabilistic theory with real-world robotic challenges.