Michael Kaess

Associate Professor, Founder of the Robot Perception Lab Carnegie Mellon University

  • Pittsburgh PA

Perception is a fundamental challenge for mobile robots navigating through and interacting with their environment.

Contact

Carnegie Mellon University

View more experts managed by Carnegie Mellon University

View all Experts

Biography

I am interested in mobile robot autonomy. One of the first problems encountered when robots operate outside controlled factory and research environments is the need to perceive their surroundings. My research focuses on efficient inference at the connection of linear algebra and probabilistic graphical models for 3D mapping and localization using information from any available sensor, including vision, laser, inertial, GPS and sonar (underwater). To enable online operation, my research also explores novel algorithms for efficient and robust inference at the intersection of linear algebra and probabilistic graphical models.

I have previously been a Research Scientist and a Postdoctoral Associate at the Massachusetts Institute of Technology (MIT), in John Leonard's Marine Robotics Lab. In 2008 I have received my PhD in Computer Science from the Georgia Institute of Technology, advised by Frank Dellaert.

Areas of Expertise

Robotics Foundations
Field & Service Robotics
Robot Navigation
Multi-Robot Systems
Autonomous Robots
Robot Motion Planning
Aerial Robotics
Underwater Robotics

Social

Accomplishments

Awards

https://www.cs.cmu.edu/~kaess/MichaelKaess_CV.pdf

Education

Georgie Institute of Technology

Ph.D.

Computer Science

2008

Georgie Institute of Technology

M.S.

Computer Science

2002

University of Karlsruhe

B.S.

Computer Science (Vordiplom Informatik)

1998

Articles

Z-Splat: Z-Axis Gaussian Splatting for Camera-Sonar Fusion

IEEE Trans. on Pattern Analysis and Machine Intelligence (PAMI)

Z. Qu, O. Vengurlekar, M. Qadri, K. Zhang, M. Kaess, C. Metzler, S. Jayasuriya, and A. Pediredla

2025-09-09

Differentiable 3D-Gaussian splatting (GS) is emerging as a prominent technique in computer vision and graphics for
reconstructing 3D scenes.

View more

“Neural feels with neural fields: Visuo-tactile perception for in-hand manipulation,”

AAAS Science Robotics

S. Suresh, H. Qi, T. Wu, T. Fan, L. Pineda, M. Lambeta, J. Malik, M. Kalakrishnan, R. Calandra, M. Kaess, J. Ortiz, and M. Mukadam

2024-11-13

To achieve human-level dexterity, robots must infer spatial awareness from multimodal sensing to reason over contact interactions.

View more

Robust Preintegrated Wheel Odometry for Off-Road Autonomous Ground Vehicles

IEEE Robotics and Automation Letters

E. Potokar, D. McGann, and M. Kaess

2024-11-11

Wheel odometry is not often used in state estimation for off-road vehicles due to frequent wheel slippage, varying wheel radii, and the 3D motion of the vehicle not fitting with the 2D nature of integrated wheel odometry. This letter attempts to overcome these issues by proposing a novel 3D preintegration of wheel encoder measurements on manifold.

View more

Powered by