Related Experiment Video
Updated: Dec 20, 2025
08:34
Automated Charting of the Visual Space of Housefly Compound Eyes
Published on: March 31, 2022
2.2K
Design and Analysis of a Single-Camera Omnistereo Sensor for Quadrotor Micro Aerial Vehicles (MAVs)
Carlos Jaramillo1, Roberto G Valenti2, Ling Guo3
1Deptartment of Computer Science, The Graduate Center, The City University of New York (CUNY), 365 Fifth Avenue, New York, NY 10016, USA. cjaramillo@gradcenter.cuny.edu.
Sensors (Basel, Switzerland)
|February 11, 2016
Summary
We developed a compact omnidirectional stereo (omnistereo) vision system for Micro Aerial Vehicles (MAVs). This system enables 3D perception from a single viewpoint, crucial for MAVs with limited payload capacity.
Area of Science:
- Robotics and Computer Vision
- Sensor Design and Performance Analysis
Background:
- Micro Aerial Vehicles (MAVs) require compact and efficient 3D perception systems.
- Existing systems may not meet the payload and performance needs of MAVs.
Purpose of the Study:
- To design and evaluate an omnidirectional stereo (omnistereo) vision system for MAVs.
- To achieve compact, 360-degree 3D sensing capabilities for MAVs.
Main Methods:
- Utilized a monocular camera with co-axially aligned hyperboloidal mirrors (vertically-folded catadioptric configuration).
- Derived analytical solutions for projective geometry based on the single viewpoint (SVP) constraint.
- Developed a probabilistic model for 3D uncertainty estimation.
Main Results:
- Demonstrated a compact omnistereo sensor suitable for MAVs.
- Generated SVP-compliant panoramic images for synchronous stereo correspondence.
- Validated projection error using synthetic and real-world data, showing 3D point clouds.
Conclusions:
- The proposed omnistereo system offers a viable solution for 3D perception on MAVs.
- The sensor design is reproducible and adaptable for various catadioptric omnistereo applications.
