RAVG: Robotics Active VisionResearch
Active Vision and Field Robotics
The research is focused primarily on developing a clear understanding of the computational and engineering aspects related to perception, reconstruction, recognition and tracking with the aim of improving existing robotic systems and/or designing new models to allow for a semi-autonomous or fully autonomous operation.
A key aspect is the use and understanding how sensor data (e.g., visual, inertial, force sensors) can be integrated in an efficient way to achieve our goals.
The research lines we address are: visual-based spatial (absolute) and topological (qualitative) localization, cognitive learning skills for navigation, active perception and human-robot interaction, visual based autonomous navigation, obstacle avoidance, 3D environment reconstruction for mobile robots, characterization of the 3D structure of environments using statistical and probabilistic methods.
Shape analysis and 3D reconstruction
This research is related with the discovery of hidden relationships in data containing a large amount of variables in comparison with a small number of observations. This falls in the field of statistical analysis with an emphasis on dimensionality reduction for building a more suitable representation for the data. Besides other applications such as humanoid robotics, acoustics and more recently agricultural analysis, we have applied dimensionality reduction techniques in two main areas of computer vision: Shape Analysis and 3D reconstruction.
Applied Computer Vision
The main topics here are related with object recognition and image understanding applied to mobile and aerial robots. We explore Machine Learning Techniques to solve real world problems.
The specific topics are:
- Image processing
- Object detection (2D and 3D)
- Computer vision in agriculture
- Drones and computer visión
- Real world perception of mobile robots
- Deep Learning for agriculture
Vision in Robotics
The research areas are:
- Motion generation and control for anthropomorphic mechanisms
- Humanoid robots
- Virtual mannequins
- Redundant arms and hands
- Vision-based robot localization and navigation
- Computational models of human walking
- Differential systems
- Statistical models