Home

The laboratory
Robots and equipment
How to reach us

 

The Laboratory

robot_thinker

 

ALCOR  has been established in 1998.

The research activity focuses mainly on the perceptual inference processes a cognitive robotics system should be able to prompt, in order to develop its knowledge about the environment and act as a mindful system.

We work both on the perceptual inference processes and on the control and planning methods based on perception.  The objective is to deal with many kind of conditions: from factory settings to harsh and unstructured settings, such as post disaster areas.

On the perceptual side we study and develop both representation and inference methods related to recognition. The recognition process is investigated as an effective inference process that informs and determines the formation of knowledge about the environment.

We build our algorithms under two approaches:  bottom-up attention and 3D reconstruction. The attention based approach starts from early attention. Under this perspective we study many aspects of visual salience, as induced by motion, by actions, and by interactions.  We are also interested in determining how classification of the different aspects of saliency can lead to the recognition of different patterns of the human behavior.

The 3D reconstruction based approach deals with both objects and scenes. 3D reconstruction of articulated objects challenges the robot knowledge formation from both its surrounding and its mental memory: the web.  3D reconstruction of very large scale areas is used by the Gaze Machine localization and also for understanding the interplay between 3D representation, saliency and recognition. In fact, the reconstruction of large areas, combined with the Gaze machine allows us to understand how people act in their habitat to accomplish simple tasks, like searching, while walking, running, crossing the road, shopping, or helping other people.

On the side of activities planning, we deal with many levels of abstraction. Our goal here is to model the knowledge flux from early perception to the representation of how choices are made. Here we start with the traversability analysis, which is the basis of motion planning in a full 3D setting. We have developed algorithms for our robot to make it able to move in an outdoor settings with many height and shape differences. This has required to combine terrain awareness together with motion control. 3D robot planning, in fact, hinders on a full comprehension of how to comply with very basic tasks in order to generate more conscious goals. These, at the current level of developments are objectives such as exploring an area aiding a rescuer or doing so collaborating with other robots or directly helping someone, by handling tools. To improve these tasks we are learning how to ensure (for our tracked robot) all locomotion abilities. So far we have focused on exploration of disaster areas, interacting with rescuers. We are now working more on collaboration with both people and other robots.

Our research is currently funded by the EU unit Robotics. We have been funded also by the Italian Space Agency ASI and by the Finmeccanica, Selex.

We are members of euRobotics aisbl since 2013.

The flyer of the laboratory can be found here.


 

Robots and Equipment

 

VICON MX

The Vicon MX system is quite simply the most advanced optical motion capture system available.The major components of a Vicon MX system are the cameras,the controlling hardware module, the software to analyze and present the data, and the host computer to run the software. Every Vicon MX system includes at least one MX Giganet to provide power and data communication with up to 10 cameras and other devices. The VICON system has been provided by GPEM S.r.l.

 

BLUEBOTICS ABSOLEM WITH KINOVA JACO ARM

ABSOLEM

Absolem is an actively articulated tracked vehicle, initially designed by Bluebotics and then, at a later stage, re-factored by Neovision s.r.o. This robot is endowed with a breakable passive differential system and two active sub-tracks, namely flippers, on both the ends of the tracks. The active flippers enhance climbing capabilities of the robot, while the mechanical differential increases the robot traction. The mechanical design maximizes the surface of the tracks in contact with the ground and improves the stability of such robot on stairs, ramps, rubbles and uneven terrain. The robotic platform is equipped with a rotating 2D SICK LMS-100 range finder, a LadyBug3 Omnidirational camera and an XSens IMU/GPS inertial sensor unit. On top of the robot body, the lightweight 6DOFs Kinova Jaco Arm has been mounted for manipulation tasks. Moreover, an IDS camera pair on a Pan-Tilt Unit-D46-17, by FLIR, enhances terrain perception, morphological adaptation and navigation.

 

THE GAZE MACHINE


The Gaze Machine (Patent number rm2007a000526: Gaze machine) is a head mounted device which allows to estimate at each instant, where the person who wears it look, in the 3D environment. It is composed of four cameras. Two high speed infrared cameras are used to track the pupils, by using specialized software. The other two cameras are pointing towards the scene that the viewer sees. These are used to provide a reconstruction of the scene. As a result, using specialized software modules, the positions where the user has looked are recovered both in the images of the scene and its 3D reconstruction.

 

SECURO Agent (SHRIMP)

BlueBotics Shrimp III mechanical architecture provides the robot with an incredible mobility. It is able to easily move in a very challenging terrain, overcome vertical obstacles of twice its wheel size, and it can even climb stairs. The additional payload is composed by a light laptop, sensor for localization (inertial platform Crossbow) and for image acquisition (two firewire cameras PtGrey Flea), a pack of Li-Ion battery, communication (Bluetooth and Wireless 802.11a). All the additional payload is mounted on ad-hoc structure made of carbon fiber and plexiglass.

 

 

DORO Agent1 (Pioneer 3DX)

 

DORO Agent2 (Pioneer P3-AT)

ActivMedia PIONEER 3-DX is an agile, versatile, intelligent mobile robotic platform able to carry loads robustly and to traverse sills, with high-performance management to provide power whenever it is needed. It is endowed with: a mobile head with a couple of stereo cameras provided by Allied Vison Technologies Marlin, an inertial platform Xsens MT9 for localization, eight sonars ring Polaroid. Other components are: a laptop Asus M3000N (centrino), a laser DISTO Leica, Odissey battery and correspondent power supply and a support structure. The software component installed on the laptop are: API ARIA and Saphira provided by ActiveMedia, Matlab (including Image Acquisition, Image Processing, Neural Network, Symbolic Math, Statistics, Optimization, Wavelet, Signal Processing toolbox), compiler C++, prolog ECLIPSE framework, PROLOG interpreter, Java. The value of the robot includes purposely developed software for localization, map reconstruction, communication and planning, navigation and recognition.


 

How to reach us

JavaScript must be enabled in order for you to use Google Maps.
However, it seems JavaScript is either disabled or not supported by your browser.
To view Google Maps, enable JavaScript by changing your browser options, and then try again.

ALCOR - Cognitive Robotics Laboratory at Sapienza University of Rome

Subscribe to me on YouTube

 
© 2017 Alcor
Joomla! is Free Software released under the GNU General Public License.