Andrew I. Comport

About

Andrew I. Comport is a computer vision and robotics researcher whose work spans visual odometry, model-based tracking, humanoid robotics, and autonomous navigation. His research has made significant contributions to the development of robust real-time systems that enable robots to perceive and interact with complex environments. Comport's early work focused on advancing visual servoing and 3D model-based tracking, with his 2005 paper "Robust model-based tracking for robot vision" garnering 91 citations and establishing efficient algorithms for monocular pose estimation critical to robot vision systems. His 2011 contribution on Direct Iterative Closest Point for RGB-D visual odometry (91 citations) demonstrated how depth-enhanced sensing could dramatically improve real-time camera motion estimation — a cornerstone technique in modern autonomous systems. Beyond foundational perception, Comport has pushed robotics into demanding real-world applications. His 2019 work on deploying humanoid robots in Airbus aircraft manufacturing (94 citations) represents a landmark achievement, bridging advanced motion planning with industrial necessity. He has also explored brain-computer interface control of humanoid robots and urban navigation using spherical representations, reflecting a remarkably broad research vision. Collectively, his publications demonstrate a career dedicated to making intelligent robots more capable, reliable, and practically deployable.

Research Focus

Key Achievements

Top Papers

1. 1
   Humanoid Robots in Aircraft Manufacturing: The Airbus Use Cases

   94 citations · 2019
2. 2
   Direct Iterative Closest Point for real-time visual odometry

   91 citations · 2011
3. 3
   Robust model-based tracking for robot vision

   91 citations · 2005
4. 4
   A spherical robot-centered representation for urban navigation

   45 citations · 2010
5. 5
   Closed-loop MPC with Dense Visual SLAM - Stability through Reactive Stepping

   19 citations · 2019
6. 6
   Object-Based Visual 3D Tracking of Articulated Objects via Kinematic Sets

   16 citations · 2005
7. 7
   Improvements in robust 2D visual servoing

   14 citations · 2004
8. 8
   Navigation assistance for a BCI-controlled humanoid robot

   13 citations · 2014
9. 9
   Efficient model-based tracking for robot vision

   13 citations · 2005
10. 10
    Appearance-based SLAM relying on a hybrid laser/omnidirectional sensor

    12 citations · 2010

Key Collaborators

Contact & Links