- Offer Profile
- Brescia University Advanced
Robotics Laboratory (ARL) was established in 1993, with the main goal of
providing researchers and students with an efficient and flexible tool for
experimental verification of ongoing research in the field of robotics.
ARL is directed by Professor Riccardo Cassinis, who has been conducting
research in the area of robotics for the last 25 years, and has founded
similar laboratories at other universities.
ARL is available to any researchers and students who need an experimental
support to their theories.
Product Portfolio
SAURON
- Costliness is one of the key problems with today’s
robots.
Cost-effectiveness is central to the SAURON (Surveillance Autonomous Robot
Over Network) project.
Developing affordable
robots, using off-the-shelf technologies and taking advantage of
infrastructures that already exist in the robot’s operating environment.
SAURON project aims at building inexpensive robots that have a moderate
interaction with the environment, such as surveillance an environmental
monitoring robots. The SAURON project focuses on development of an autonomous vehicle for
indoor and outdoor use, capable of carrying video cameras and other sensors
to provide a real time link with surveillance posts or data collection
facilities.
The key features of the project include:
• Extensive use of existing components, acquired as
much as possible from the consumer market, thus keeping the robot’s cost as
low as possible
•
Intensive use of resources already present in the environment (sensors, data
communication net-works, public phone services, computing resources, etc.)
•
Development of more economical technology for self-localization and mapping,
using less expensive components, thus further reducing total cost of
to-day’s robots MORGUL
MORGUL (Mobile Observation Robot for Guarding the University Laboratories)
robot, in its current version, is an ActivMedia Pioneer 3AT robot equipped
with:
• Onboard web-cam
• Active marker
• Laptop
• APNS (on laptop)
• ARIA robot programming libraries (on laptop)
• Wireless networking devices
• Sonar belt
• Front and rear bumpers
MORGUL also has a docking station, that can be reached using a simple
optical homing system that uses the onboard webcam. The docking station
contains a battery charging system, and allows fully autonomous operation of
the robot over extended periods of time.
AMIRoLoS
Localization is one of the fundamental problems in mobile robot navigation.
AMIRoLoS (Active Marker Internet-based Robot Localization System) implements
a new methodology aimed at localizing a mobile robot in indoor and outdoor
environments using active markers and commercial off-the-shelf webcams. The
marker detection system, which is based on the difference of working
frequencies of the shutter of a webcam and of a signal from the marker, can
be detected at a distance of up to 100m from the camera in full sunlight.
APNS
APNS (Automatic Predictive Network Selection) is a software package designed
to allow a mobile robot to autonomously manage the access to heterogeneous
wireless TCP/IP networks. With this software, the robot is able to look for
an available communication channel and to establish a new connection to it.
Moreover, if the robot detects more than one network, it can choose the best
one according to predefined criteria. The main goal of this project is to
make the robot able to use, always and everywhere, the best available way of
communicating, without the need of explicitly connecting end disconnecting
from available networks.
This task can be completed without any awareness of the context in which the
robot is: this means that the robot can immediately start working in a new
location with no need for reconfiguration. However, the system can take
advantage of previously acquired knowledge to choose in advance the most
appropriate connection. In a future version, knowledge of the task the robot
is performing will also be used to optimize communication.
ARCSUS
(Advanced Robot Control System Using SVG)
In order to visualize and control the system, we are developing an SVG
graphic interface that will allow controlling the system from local or
remote computers using standard browsers. The interface can be visualized
over the Internet by any browser and any OS with an SVG plug-in using W3C
standards (W3C validation in XHTML 1.0 and CSS).
De-mining Robot Project
- The highly expensive (in economical terms and, much
more important, for its human lives cost) problem of cleaning mines from
former war areas is one of the most challenging for robotics engineers.
As mines were obviously built to be undetectable, locating them is a highly
dangerous and difficult task. Many worldwide projects were started, some
concentrating on a single kind of sensor, some using several different
sensors and integrating the information they provide. Our project, carried
on in cooperation with researchers from the Chemistry and Physics Department
of our University, focuses on detecting the odors of chemical substances
associated with mines. This requires highly sensitive gas detectors, and,
maybe even more important, an intelligent search strategy that will take
into account environmental conditions, wind speed and direction, etc. We are
fairly convinced that no single sensor will solve the problem, but multiple
sensors (or multiple single-sensor robots operating in a coordinated way)
can attain a very high level of reliability.
An international research consortium is being organized to apply for funding
for this research. Sensors & Robots for demining
Sensors:
During the last few years, dozens of other methods have been developed for
detecting landmines. Each of them, however, has advantages and drawbacks,
and no method exists that can be regarded to as universal. We have compared
all existing methods (at least, all the methods we know) in a single table.
Robots:
In order to understand what is now available on the market and in research
laboratories as far as robots for demining are concerned, we have carried on
a survey about existing machines. The results, along with pictures and some
movies, are shown here.
Mine Sniffer Project
- As mines were obviously built to be undetectable,
locating them is a highly dangerous and difficult task. Many world-wide
projects were started, some concentrating on a single kind of sensor, some
using several different sensors and integrating the information they
provide. This project, carried on in cooperation with researchers from the
Chemistry and Physics Department of our University, focuses on detecting the
odors of chemical substances associated with mines. This requires highly
sensitive gas detectors, and, maybe even more important, an intelligent
search strategy that will take into account environmental conditions, wind
speed and direction, etc. To demonstrate the feasibility of the project, two
prototypes have been built, that can locate odor sources using and
appropriate strategy.