Mikko Lauri
Papers
8
Total Citations
351
H-Index
6
About
Mikko Lauri is a leading researcher at the intersection of robotics, decision-making under uncertainty, and autonomous information gathering. His work is unified by a focus on Partially Observable Markov Decision Processes (POMDPs), which provide a principled framework for robots operating with noisy sensors and imperfect control. His landmark survey, "Partially Observable Markov Decision Processes in Robotics: A Survey" (2022, 150 citations), has become a foundational reference for the field, systematically mapping how POMDPs can address real-world challenges in robot autonomy. Lauri has made significant contributions to active perception and multi-robot coordination. His work on "Planning for robotic exploration based on forward simulation" (2016, 101 citations) introduced efficient simulation-based methods for autonomous exploration. He has also advanced multi-sensor planning, formulating the next-best-view problem for robot teams as matroid-constrained submodular maximization (2020, 36 citations), and developed strategies for multi-robot information gathering under periodic communication constraints (2017). His research on occlusion-resistant object rotation regression from point clouds (2019, 23 citations) and deep reinforcement learning for active visual object search (2019) further demonstrates his impact on practical robotic perception and manipulation.
Research Focus
Key Achievements
Top Papers
- 1Partially Observable Markov Decision Processes in Robotics: A Survey150 citations · 2022
- 2Planning for robotic exploration based on forward simulation101 citations · 2016
- 3
- 4Multi-robot active information gathering with periodic communication23 citations · 2017
- 5Occlusion Resistant Object Rotation Regression from Point Cloud Segments23 citations · 2019
- 6
- 7Myopic Policy Bounds for Information Acquisition POMDPs2 citations · 2016
- 8Occlusion Resistant Object Rotation Regression from Point Cloud Segments2 citations · 2018