Related Experiment Video
Updated: Jan 3, 2026
07:45
Assessing Binocular Central Visual Field and Binocular Eye Movements in a Dichoptic Viewing Condition
Published on: July 21, 2020
4.9K
Sliding Window Mapping for Omnidirectional RGB-D Sensors.
Nicolas Dalmedico1, Marco Antônio Simões Teixeira1, Higor Barbosa Santos1
1Graduate Program in Electrical and Computer Engineering (CPGEI), Universidade Tecnológica Federal do Paraná (UTFPR), Curitiba PR 80230-901, Brazil.
Sensors (Basel, Switzerland)
|November 27, 2019
Summary
This study introduces an omnidirectional RGB-D sensor prototype with a novel mapping strategy for robots. It enables 360-degree environment perception and reconstruction for tasks like human detection.
Area of Science:
- Robotics
- Computer Vision
- Sensor Technology
Background:
- Robots require comprehensive environmental perception for navigation and interaction.
- Existing sensors often have limited fields of view, necessitating complex configurations for omnidirectional sensing.
Purpose of the Study:
- To develop an omnidirectional RGB-D sensor prototype.
- To create a novel mapping strategy tailored to the sensor's capabilities.
- To enable robust environment reconstruction and human detection for mobile robots.
Main Methods:
- An actuated Light Detection and Ranging (LIDAR) sensor combined with an RGB camera.
- A 90-degree tilting mechanism and rotational capability for the LIDAR sensor to achieve omnidirectional scanning.
- A dual-map strategy: a local 2D map for immediate perception and a global map for memory.
Main Results:
- The prototype successfully gathers RGB and 3D data from all directions.
- The proposed mapping strategy effectively utilizes sensor data for environment reconstruction.
- The system demonstrates potential for accurate human detection within the robot's environment.
Conclusions:
- The developed omnidirectional RGB-D sensor and mapping strategy enhance robot perception capabilities.
- This approach offers a flexible and effective solution for 3D environment mapping and analysis.
- The system provides a foundation for advanced robotic applications requiring full-scene awareness.
