Andrea Censi

About

Andrea Censi is a robotics and computer vision researcher whose work spans event-based sensing, robot localization, calibration, and autonomous systems. Perhaps his most influential contribution is his comprehensive survey on event-based vision (2020, 633 citations), which helped establish the theoretical and practical foundations for this emerging field. Event cameras — bio-inspired sensors that asynchronously capture per-pixel brightness changes rather than fixed-rate frames — have become an exciting frontier in robotics, and Censi's work has been central to shaping how researchers approach them. His pioneering investigations into low-latency event-based visual odometry and LED marker tracking demonstrated that dynamic vision sensors could dramatically reduce processing pipeline latency, enabling more agile autonomous robots. Censi has also made lasting contributions to classical robotics problems: his simultaneous calibration methods for odometry and sensor parameters (collectively exceeding 170 citations) provide elegant, practical solutions for mobile robot setup. His scan matching work in the Hough domain advanced SLAM techniques, while his research on robust pose graph optimization introduced convex relaxation approaches to handle real-world measurement outliers. Beyond algorithms, his OpenRDK framework supported modular robotic software development. Across domains from microscale flying robots to probabilistic motion planning, Censi's research consistently bridges theoretical rigor with practical robotic implementation.

Research Focus

Key Achievements

Top Papers

1. 1
   Event-Based Vision: A Survey

   633 citations · 2020
2. 2
   Low-latency event-based visual odometry

   131 citations · 2014
3. 3
   Simultaneous Calibration of Odometry and Sensor Parameters for Mobile Robots

   121 citations · 2013
4. 4
   Controlling free flight of a robotic fly using an onboard vision sensor inspired by insect ocelli

   114 citations · 2014
5. 5
   Scan Matching in the Hough Domain

   105 citations · 2006
6. 6
   Low-latency localization by active LED markers tracking using a dynamic vision sensor

   101 citations · 2013
7. 7
   Selecting good measurements via ℓ<inf>1</inf> relaxation: A convex approach for robust estimation over graphs

   82 citations · 2014
8. 8
   OpenRDK: A modular framework for robotic software development

   79 citations · 2008
9. 9
   Simultaneous maximum-likelihood calibration of odometry and sensor parameters

   51 citations · 2008
10. 10
    A Bayesian framework for optimal motion planning with uncertainty

    51 citations · 2008

Key Collaborators

Contact & Links