well
uh i will try to be fast because there is already select for you right
so
it's my pleasure to go down a introduce a this project through this is european
project uh unique properties again
uh the same the project is robust and save mobile cooperation of the number sixteens
and the abbreviation is a actually
uh so at the beginning and introduce the project itself in schools structure and then
demonstrators
oh then i will present our scientific development and achievements in this work or university
and finally i will describe how we integrate the our results into the project
the ministry
so the main project all its production of advanced robust and C cognitive uh reasoning
older than a simple practical robotic systems at reduced cost is most important thing
then designed and it'll reusable building blocks collected in the knowledge base
is the main idea of the project that
knowledge base with
two rolls independent building blocks that are reusable so then any other repeatable will be
cheaper in
present time
so based on the state-of-the-art that is
uh the project goals is to specified objects and develop in about solutions covering or
crucial robotics topics
uh such as however performance in addressing and i'm and perception
and modeling reasoning decision making or validation and testing
that and design of solutions are used to develop a knowledge base
uh i and tools for body development and standardised testing
a button on to manage those two approach
one of the key or what
as i said is development knowledge base but for knowledge based framework
so first two points
state-of-the-art records
and new features are used to feel this knowledge base
and used to design a methodology i'll work with this knowledge
the model
solutions will be then
integrated in the demonstrators from a different industrial partners
so there are examples of application domains
they are covered by industrial partners in this project
um
maybe i was giving you to show you the partners that is twenty seven partners
in position quite large amount
the project is focused on corporation this industrial partners so besides universities and national research
centres there is a high part oh
members of the project
uh our industrial representatives
so come back to this light
both and that area and probably both
they are not round by
um independently or incorporation are imperative solutions for accomplishment of surveillance or security inspection you
to use a more extensive address
examples might be more ignoring the borders the seashore all inspecting infrastructures some
about exactly was i'll talk in detail later because this is what we integrate our
solutions
industrial in manufacturing wouldn't parts and products the process maybe
like cutting the who surf a single assembly of the parts that are usually carried
by nancy on C and C O machines
in the current practise
oh
manual operation with machines is necessary because the machines are not able to of
oh to the mostly work with small
parts so the objective of this project is also solve these problems and then there
is a set of this remote applications that are can ski because we california talked
about the previous
presentation in detail
so in this structure where is a more research and it is it is university
there are driven by requirements of objects applications
but online as real-time sensor data processing is key requirement
our proposal cues and development novel methods and optimization of existing ones and acceleration
in our some
the methods are well separated might be so i
that is what we have developed some methods and the results were research O R
based on says a fusion usually for those tasks for robot localisation an object detection
of perception
in the robot localisation uh we develop a method that uses two
use an existing methods is you mapping this method that process a laser scans and
have very low precision but the model as you can see the environment is quite
or so
it's harder to get any knowledge from it
other methods
based on data from kinect so we show and that's data
can create a more informative model but the precision is so matter if use those
two methods actually we are able to get much we can see the precision of
a methods domain
and so
one remote so
final solution is um precise and the model was then higher quality
including detection
we experimented and realised several methods i will briefly introduced to them in that um
when we achieved a nice results that the uh published
well all those detectors are then used in our system and fused together well for
improving the robustness the final solution and the
more precise modeling of the environment
situation so one that is
a segmentation of the video signal um by tracking of local features
in comparison with existing methods that are used more on
precision and instability of the method as i said i focus on
speech and also
to work in an online manner
but usually off-line methods that exists use whole data to produce minus tracks and then
cost in a nice way we cannot use all data from
the future in on-line processing so this work this was uh problems that we need
to solve so actually come up with the
the method is able to run online and the real time with comparable precision but
with
many more times higher uh speech so
computational cost is very low
this method is able to segment one the robot moves
this method is able to segment objects that moves in the video stream relatively to
each other
so it doesn't necessarily mean that the T V to move but if there is
object far from some background we are able to see actually i shouldn't say we
this is computed at the picture this is a segmentation you don't know if it's
object one C
the second method i want to introduce is um
processing methods for this data um
the idea is based
that
in indoor scenes many objects or wonder so
this method segments longer well
segments object
again before used on a computational efficiency so here we have
also slightly better precision than existing methods but there are many times faster than the
others
this is the last uh and rough research what we do here for the project
this is for validation and verification part of the project reading something we should of
common uh features
um
usually one
robin systems are the lot and need to be very verify that there is need
of some simulation process
and usually we generate uh image actually
right image so what we are trying to do is to somehow similar in the
real situation so
we in
introducing to be nice um
say
distortions
right lower noise and many other
distortion
chromatic aberration
lens flare to basically a example
not perfect
so that was our research from scientific point of you know how we apply an
integrated or assumptions into a vectors we cooperate with i hope the at like a
different at that is um
uh
they have a covert in for gifts those allergies laser guided very close
just physically moving groups
and X is a link between different machines in big barrels moving the pilots and
they do it autonomously controlled by one
uh
central unit
i that controls oh
situation in the better so we have sold to task with that
one is also abundance this is crucial for that a lot
oh
solution because when two i T Vs meets somewhere because of one hundred you can
feel because of something like models and yeah anyway however than rgb space there and
in the corridor but i don't actually use
so we need to solve all the other actually might
and white this obstacle
the second
task is um
this application track probably means that those ics needs to get into that right lower
their the product and the track the problem is that
the solution
proposition solution for this uh framework
is based on there is an allegation positioning system that works only inside the white
house indulgently the rubber believes the barrels it was is position is as see what
it is so the goal is somehow
measure and perceive
the localisation actually out of the barrels inside the track the track which later there's
also some constraints
so this to make that we try to uh
so
this example this is serious frames rubber problems sensor is a camera
this is manually driven
example how
and we should be hey if i think it's J
and then we see what type do this
it's a person if it so that the robot it's politics
this an object because according to those information we can say if you can what
do not because of the safety reasons for example the person is not
ll to be avoided
the attendant actually or pilot is about the word in the harness and so
uh so
for this we design some structure i would represent what here and its input data
centre we use of actions in some sort
in all image domain or sensory that my
our experiments we use
R G and G other constraints bottom an environment we
right project and what you don't objects in the environment
alright
having
information about objects don't actually are report classify what type it is if it's a
danger situation just warning situation
a point of the information that rgb in making decide if he wants to avoid
object so that it is that we have a planning module that computes the past
and uh provide information twenty actually here we do the perception and more than the
a
i don't know any controlling on T V
so we only provide measurements and
that's a
oh analysis and proposals what actually my too much oh but you know any controlling
is it an example
just of the constraints in front of the T that there is something classified object
read position study danger or if it's just morning andrea
according to those information we use something like decision making use a if it's a
person it is too close to be muscle or the bill if it's other robot
and it's far all what we need to do is slow down so this is
people describing the decision making and remind you also computes
either that a if the avoidance
procedure is exceeded
so this is our asepsis team for obstacle avoidance cost
and the second that it is it is to get into the track
localise itself in the strong using different localisation method then used in our house so
you can see
i wanted visual information track is
to do we base or solution or laser measurements
so we design simply the model of that right and we measured roles right points
uh in that right
and we provide measurements
to the T V and T V then using a similar way as uh it's
postage it's uh it's using the localisation information
one house system
so
we have developed a methods
for sensory processing
i think the next imputation
we use an optimization method one
acceleration hardware
um but it somewhere else in which can directors oh
the results of those methods are was rendered looks like the use of two hours
we created experiment experimental probably form that see here our experiments
oh really
and most of our results are great to demonstrate
these cases
um thank you very much attention