|
|
European
|
||||
|
|
||||
|
|
||||
|
|
||||
|
|||||
|
Research Projects (under contract)
|
|
European
|
||||
|
|
||||
|
|
||||
|
|
||||
|
|||||
|
National
Projects
VISUAL BEHAVIORS FOR MOBILE ROBOTS
JNICT PBIC/C/TPR/2550/95
Jan 1996-Dec 1998
Project Objectives
One of the major limitations of the current robotics systems is their reduced capabilities of perceiving the surrounding space. This limitation determines the maximum complexity of the tasks they may perform, and reduces the robustness when the current tasks are performed.
Therefore, by increasing the perceptual capabilities, these systems could react to environmental changes, and accomplish the desired tasks. For many living species, namely the human being, visual perception plays a key role in their behavior. Very often we rely intensively on our visual capabilities to move around in the world, track moving objects, handle tools, avoid obstacles, etc.
To improve the flexibility and robustness of robotics systems, this project aims at studying and implementing Computer Vision techniques in various tasks for Mobile Robotic Systems. The goal is to study not only the visual perception techniques, per si, but also to explore the intimate relationship between perception and the control of action : the Perception-Action cycle.
For many years, most research efforts on Computer Vision for Robotic Agents were focused on recovering a simbolic model of the surrounding environment. This model could then be used by higher level cognitive systems to plan the actions for the agent, based on the pursued goals and the world state. This approach, however, has revealed many problems in dealing with dynamic environments where unpredictable events may occur.
More recent approaches, trying to achieve robust operation in dynamic, weakly structured environments, consider a set of behaviours, in which perception (vision in this case) and action are tightly connected and mutually constraining, similarly to many successful biological systems. Therefore visual information is fed directly into the various control systems of the different behaviours, thus leading to a more robust performance.
Specifically, in this project, an Architecture for Visual Behaviours for a mobile vehicle equipped with an agile camera, will be considered. Each behaviour allocates the perception and action resources strictly needed for the associated task, such as detecting and tracking a moving target, detecting interesting points in the scene, docking to a specific point, detecting obstacles, navigating along corridors, self- localization, etc. The robustness and performance of the overall system emerges from the coordination, integration and competition between these various visual behaviours.
SIVA - Integrated System of Active Surveillance
PRAXIS 2/2.1/TPAR/2074/95
Jan 1998-April 2001
Project Objectives
Research Areas: Computer Vision,Mobile Robotics
External partners: ISR-Coimbra Pole
With this project we aim to develop an autonomous system able to perform
surveillance tasks in a structured environment. The system as a whole will be
made up of several mobile platforms (“agents”). Each of them will be
equipped with an active vision system, an odometry system and communication
links enabling the platforms to communicate among themselves. As a whole the
full system (including all the platforms) will have a high degree of autonomy.
Each of the platforms will also be autonomous in terms of the navigation and
surveillance tasks. The system will be designed to operate off-hours in
structured environments such as supermarkets, military installations, public
buildings, power plants, etc.. The implication of its use during off-hours is
that the primary condition for the detection of an intruder will be the
occurrence of motion. Each one of
the “agents” will perform the surveillance tasks based on the active vision
system. Navigation will also be performed based on the active vision system and
on the odometry system. The environment map will be known “a priori” and
landmarks both natural and artificial will be used for localization. Both
navigation and surveillance do not have to be performed with accuracy. Indeed
each “agent” will only be concerned with not repeating its trajectory so
that all the map area is covered periodically. Agents will have to cooperate so
that they do not survey the same area simultaneously. Therefore there will be
communication links enabling the “agents” to inform one another from their
activities. Cooperation will occur at several levels namely to ensure that the
“agents” survey different areas, to avoid collisions and to enable the
execution of common taks such as pursuing an intruder.
We propose to demonstrate this system by using two mobile and active
“agents” that cooperate in a structured environment.
SAPIENS PROJECT 34121/99
2000-2003
Project Objectives
Research Areas: Signal Processing, Computer Vision
Classification: SAPIENS PROJECT 34121/99
This
project proposes to develop a methodology for video coding by using intermediate
3D representations In other words we propose to code a sequence ofvideo images
into a 3D representation and then
generate back another sequence. The main objective is to create a high
compression rate and make possible mecanisms for video indexing and interactive
video.
The
main topics we propose to tackle are: (1) Image to image matching, (2) image to
model matching and (3) 3D modeling and generation of images from the 3D model.
The
main ideia is to create a feedback loop of "a priori" knowlege
provided by the existent 3D scene
model into the matching process, in a global way.By formalizing image to
image and image to model matching
as an integer programming problem which is then relaxed to a concave
programming problem (see [5] ), authors
believe the whole process of matching and 3D reconstruction can be integrated
into a single recursive framework.
European Projects
NARVAL-NAVIGATION OF AUTONOMOUS ROBOTS VIA ACTIVE ENVIRONMENTAL PERCEPTION
Esprit LTR Project-30185
1998-Dec 2001
Project Objectives
Research Areas: Computer Vision,Signal Processing,Navigation
External Partners: 13S (CNRS-Université de NIce Sophia Antipolis) (FR9,Thomson Sintra ASM (FR),DIST-University of Genova (I)
The goal of this project is to develop non-intruding and reliable
navigation systems giving the robot the ability to select natural landmarks, and
to navigate with respect to them, extending in this way the autonomy range of
autonomous robots operating in unknown and unstructured environments. Reliability
is achieved by continuously controlling the uncertainty associated with
knowledge of the environment and of the robot's position and orientation. In the
context of the project, non-intrusiveness means that the robot must be
able to operate without special purpose landmarks being added to its
environment. Non-intrusive operation in unstructured environments
precludes navigation with respect to a set of handcrafted “landmarks,” and
requires the robot's ability to infer its position from learned natural
landmarks of the environment using its perception system. Instead of passive
reconstruction of the working space, perception is faced as a process of
selectively extracting from the world the information needed to accomplish a
given task, trading generality for specificity and gaining in simplicity and
robustness. It is no longer a separate off-line module, but an integral part of
the closed loop control system. This coupling will be explicitly addressed at
the control level by assessing the compatibility of the current state of the
robot's knowledge of its environment, its mission and safety requirements.
Availability of such systems has a considerable impact in many economic, social
and industrial activities such as control of marine pollution, surveillance of
restricted areas, surveillance of equipment, agriculture, underwater cartography
and marine biology studies, to mention but a few.
VIRSBS-VISUAL INTELLIGENT RECOGNITION FOR SECURE BANKING SERVICES
Esprit LTR-Proj.21894
1996-Feb.1999
Project Objectives
Research Areas: Computer Vision,Signal Processing
External partners: ISR-DIST-Universitŕ di Genova (coordinator-I),École Polytechnique Féderale de Lausanne (CH),Maynouth College
The goal of this Reactive LTR project is to realize a prototypeautonomous station for personal identification. This station will include all the features required to be integrated in to a newgeneration of automated security check-point along corridors,passageways or access doors, and in the next-generation ofautomatic teller machines. This prototype will be used to performa significant set of statistical test on personal identification.Secure access control is a key issue in banking services. Magneticcards and personal identification numbers, currently adopted foraccessing automatic tellers, do not provide a sufficient degree ofsecurity and are likely a source of unauthorized operations. Asfar as the access to restricted areas is concerned, it usuallyrequires direct surveillance by guards or indirect surveillance bya human operator through a monitoring system. Even in this case it is often difficult, due to fatigue or other distracting factors,to guarantee continuous and high performance in this task.
The project is mainly focused on
banking services, in particular for secure and safe control of access to key
areas in the bank building and for cross-checking personal identity of peoplerequesting banking transactions. A system based on visual
recognition will have a major impact on man machine interaction, providing a more
natural way for the customer to interact with the banking security system. The project will also have a potential
impact in various scenarios ranging from generic surveillance in buildings and parking areas to
security control through check points in airports or railway stations.
Our approach will be to exploit newly-developed advanced techniques based on computer vision and robotics. Iconic and
feature-based techniques will be utilized in the first instance and in case of ambiguities performances will be improved by using
stereo analysis and the general theory of projective invariants. A
breakthrough with respect to current technology will be given by the use of space-variant image
representation and by the realization of an active robotic system, able to fixate and trackthe examined subject.
VIRTUOUS-AUTONOMOUS ACQUISITION OF VIRTUAL REALITY MODELS FROM REAL WORLD SCENES
Inco-Copernicus Proj. 960174
1997-1999
Project Objectives
Research Areas: Research Areas/Groups involved in the project (eg.Control Theory,Signal Processing etc.)
External partners: Centre for Vision,Speech and Signal Processing (University of Surrey,UK-coordinator),Institute of Control Theory and Robotics (Slovak Academy of Sciences,Slovakia),Institute of Information Theory and Automation (Czech Academy of Sciences,Czech Republic)
As virtual
reality (VR) systems improve in performance attention is turning towards the
content of virtual worlds and what can be done within virtual worlds. To make
virtual worlds interesting
detailed scene models must be built. The models need to contain 3D shapes as
complex as those that we are familiar with from our everyday experience. Shape
alone is not enough, the real world has an infinite variety of colours and
textures which also need to be included in virtual worlds.
Developing these models using 3D modelling software becomes more time consuming
as the scene complexity increases. The objective of this project is to capture
virtual reality (VR) models of real
world scenes and then to use these models. Within this project we will examine
two sensor systems to acquire VR models. The acquisition of single view range
images using structured light projectors is approaching commercial maturity. The
University of Surrey will investigate the use of multiple views of range images
combined with colour images to produce VR models.
The Instituto Superior Técnico / Instituto de Sistemas e Robótica will
use computer vision techniques to acquire scene models from video sequences
taken from a mobile platform. Although the objective is the same as the
University of Surrey, this is a more ambitious sensor configuration. It has
lower cost but will need more powerful software.
For a virtual world to have a satisfactory look and feel it must have
lifelike colour and texture. Not only do these provide useful cues to a subject
navigating in such an environment, but they also aid in the accurate detailed
reconstruction of the environment. The Institute of Information Theory and
Automation will address texture segmentation and synthesis. Their objectives
will include reducing the space and time overheads of texture in VR systems.
Virtual reality systems can be used for a variety of applications in
entertainment, medicine and manufacturing. It follows that producing detailed
models is of generic interest. Within this project we will address one
particular application. The main application to be developed is a Virtual
Reality Robot Arm Trainer, and will be done by the Institute of Control Theory
and Robotics (ICTR) . This will also provide a mechanism to validate the scene
models.
EU-FET-2000-28159
September 2001Project Objectives
Research Areas: Computer Vision
External Partners: DIST-University of Genova (I),University of Ferrara (I),Dept.of Psyhology University of Umea,(SE)
The goals of MIRROR are: 1) to realize an artificial system that learns to communicate with humans by means of body gestures and 2) to study the mechanisms used by the brain to learn and represent gestures. The biological base is the existence in primates’s premotor cortex of a motor resonant system, called mirror neurons, activated both during execution of goal directed actions and during observation of similar actions performed by others. This unified representation may subserve the learning of goal directed actions during development and the recognition of motor acts, when visually perceived. In MIRROR we investigate this ontogenetic pathway in two ways: 1) by realizing a system that learns to move AND to understand movements on the basis of the visually perceived motion and the associated motor commands and 2) by correlated electrophysiological experiments.
OMNIVIEWS-OMNIDIRECTIONAL VISION SYSTEM
IST-Proj. 1999-29017
September 2000-September 2001
Project Objectives
Research Areas: Computer Vision
External partners:DIST-University of Genova (I),Czech Technical University (Cz)
The goal
of the project is to integrate optical, optoelectronic, hardware, and software
technology to realise a smart visual sensor, and to demonstrate its utility in
key application areas. In particular our intention is to design and realise a
low-cost, miniaturised digital camera acquiring panoramic (360 deg) images and
performing a useful low-level processing on the incoming stream of images in
real-time. Target applications include surveillance, quality control and mobile
robot and vehicle navigation.
