Player is loading...
Towards human-centered robotics
Sylvain Calinon, Research Director at the Idiap Research Institute
Saturday, Sept. 11, 2021 · 5 p.m. · 09m 30s
Embed
Transcriptions
Note: this content has been automatically generated.
00:00:00
very much indeed for the introduction and thank you for inviting me going to start my presentation with little gain help
00:00:07
me to find where charlie is in this video it
00:00:12
it's quite a difficult exercise because in this type of application
00:00:17
for a few people so we're on the factory floor you can
00:00:21
find a few people but interaction between robots and humans is pretty limited
00:00:29
the only interaction usually is if i'm human and does the perimeter of the robot then the the machine stops
00:00:39
i know in manufacturing enterprises to use a robots which use the same
00:00:45
work spaces humans which collaborate with humans which are used as a tool
00:00:49
and we have existing robotics companies which are creating these new products and new companies which of flirting
00:00:55
and which specialise in this type of collaborative robot or
00:01:00
we also have a lot of robotics uh applications which uh i'm not in the context of for
00:01:07
manufacturing and which are closer to us robots which are close to us and further away from us
00:01:14
the ones which are close to us our system it's robots
00:01:19
for example surgical robots and those which are further away from us
00:01:26
um they often i use because it's a place which is too remote for us to go to
00:01:31
and therefore we prefer to send the robot and to pretend that we are present uh but
00:01:35
uh we replaced by robots uh uh remotely
00:01:39
this can be tele present so tele operation applications
00:01:45
what is common to these different forms of robots is that these robots
00:01:48
are required to be programmed to much more frequently than robots
00:01:52
in the automotive industry as i showed you in the video
00:01:56
so uh how a research group has been looking into the
00:02:00
way in which these robots are to be programmed and to use
00:02:03
uh imitation mechanisms to be able to do so
00:02:09
to teach the robot how to do task in exactly the same way you would teach a human to do the task
00:02:16
i will show you this in the various examples first of all with
00:02:19
the collaborative robots let's look first of all what a a robot is
00:02:25
if we take a human lloyd the robot you can define it technically as a series
00:02:30
of actually doesn't sensors and the way in which these are connected up but if you
00:02:36
define it in this way up there are many other things you could define in the same way
00:02:41
for example an automatic doors also sensor and actuator in the way of looking the two up
00:02:47
in a way the robots suffers from the fact that in the first context
00:02:53
we're certainly going to be afraid that the robot might just be replacement
00:02:58
was in the second context we're probably going to forget the fact that
00:03:04
once this mechanism
00:03:11
um mania we use to replace certain
00:03:14
jobs we can ask ourselves uh questions about
00:03:19
the the word to robot itself uh often robot uh as george if connotations
00:03:26
and some uses are going to move on to the code demand
00:03:28
project which is a a corporation with my colleagues said the burn uh
00:03:35
university of applied science is with these two people whose names you see
00:03:40
on the top right of the screen for us uh russia and professor
00:03:47
oh
00:03:50
and isn't that the idea is to do cyclical testing the
00:03:54
approaches centred on the person's we're going to use these technologies
00:04:00
and they are going to be used for different needs
00:04:06
we're talking about uh in does different industrial needs a agility
00:04:11
three different interest only to julie t. productivity inflexible team we're going to intervene on the people who really need these tools
00:04:18
yeah maybe we have an idea in mind about this
00:04:20
technology is we observe how these people will use those technologies
00:04:24
and we'll see whether they can help us to re format these technologies so that they can be even more useful
00:04:31
so the notion behind this project is one of empowerment to really see the
00:04:38
um robot is a tool which and people can use
00:04:42
in the way in which they wish maybe some people will
00:04:46
use it to very little it but this flexibility also
00:04:50
has to be provided so that they can evolve over time
00:04:57
initially the robot will be tested for task which is not really
00:05:00
important and then the people are testing it will be able to uh
00:05:06
contributor on a um knowledge to quite this technology i'm
00:05:12
going to a few words about robots used for assisting people
00:05:16
here we see a large number of new applications for assistance robots let me give you some examples
00:05:24
for example robotic and help someone to get out of bed to promote
00:05:28
that if it's your therapist robot and robot which helps people to make fruit
00:05:37
bottom left and hear a robot which can help with washing
00:05:45
okay what simple just have this flexibility aspect but also the personalisation
00:05:48
aspect in other words each person is going to have different needs
00:05:55
and of course it's also important to be able to reprogram the machine according to the person
00:06:02
often it's a tripartite uh uh interaction it's not only
00:06:06
between the and one person and the robot but maybe
00:06:09
also the medical staff who have to be able to reprogram the machine so to third party can use it
00:06:19
no and i there are two projects i would like to
00:06:23
describe the with regard to this and uh application i dress
00:06:29
i have two scenarios helping someone to put on the shoe or to put on a jacket
00:06:35
and behind the person you can see the technological elements
00:06:39
that you'd need in order to be able to help
00:06:41
this person the big chunk here is that you have to combine not only a movement but an applied force
00:06:49
so the person is not constrained by the robot but can move freely
00:06:53
so you have to have a robot which needs to be flexible most of the time
00:06:57
and rigid only when you really have to apply certain forced to to a specific task
00:07:05
another example of like to show you is a project which
00:07:09
started recently which we would like to develop a robot turn
00:07:15
to provide personalised assistance to help us and uh stand
00:07:19
up and sit down if you the model of what happens
00:07:22
if if you know the person and if you know the robot what has to be done to help this person
00:07:31
e. your seven and some hypotheses about the a number of points of contact
00:07:37
uh personalisation involve saying that rather of course joints
00:07:42
some joints which may be more problematic than others and we
00:07:46
want to provide assistance which can be adapted to this specific scenario
00:07:51
one person might uh i'm i'm joint which hurts him or her or uh which does not have enough muscles
00:07:58
the project is still in its infancy
00:08:02
the uh i use is a very important we're doing this with our colleagues from the ages of stuff and
00:08:10
uh institute in slovenia who specialise in studying behaviour
00:08:16
yeah the buttons when two people collaborate with one another but also with all the
00:08:20
bile mechanical aspects not what's being able to collect the data with several types of senses
00:08:28
you know to to understand what type of assistance you need for this
00:08:32
particular application work or the sun summary of the message i would like to
00:08:44
country with this presentation is that when you think of
00:08:47
the word robotics you have to realise that to row row
00:08:51
robotics is very fast um can be assistance robotics collaborative
00:08:55
in industry or process prostheses all rehabilitation tools and so on
00:09:03
uh will uh and with this last slide which shows you the faces of the
00:09:07
people who are behind the projects that i've just described you think if your attention
00:09:18
uh_huh
00:09:21
thank you i think that
Conference Program
Welcome words
Aurélie Rosemberg, Fondation Dalle Molle
Sept. 11, 2021 · 4 p.m.
Artificial intelligence and quality of life
H. Bourlard, Idiap Research Institute
Sept. 11, 2021 · 4:30 p.m.
Artificial intelligence to think like humans
Melanie Mitchell, Professor at the Santa Fe Institute
Sept. 11, 2021 · 4:45 p.m.
Towards human-centered robotics
Sylvain Calinon, Research Director at the Idiap Research Institute
Sept. 11, 2021 · 5 p.m.
Supporting sustainable transitions around the world through water technology
Eric Valette, Director of AQUA4D
Sept. 11, 2021 · 5:15 p.m.
Biometric security
Sébastien Marcel, Research Director at the Idiap Research Institute
Sept. 11, 2021 · 5:30 p.m.
Compatibility with humans: AI and the problem of control
Stuart russel, Professor of Computer Science and Smith-Zadeh Professor of Engineering, University of California, Honorary Fellow of Berkeley and Wadham College at Oxford
Sept. 11, 2021 · 5:45 p.m.
Model subjectivity at the heart of consciousness to make robots more human
David Rudrauf, Associate professor at the University of Geneva, Director of the laboratory of the multimodal modeling of Emotion and Feeling
Sept. 11, 2021 · 6 p.m.
Round table
Panel
Sept. 11, 2021 · 6:15 p.m.
Recommended talks
Vers la robotique centrée autour de l’humain
Sylvain Calinon, Directeur de recherche à l’Institut de Recherche Idiap
Sept. 11, 2021 · 5 p.m.
session IM2 Start-up Jamboree: Kooaba
Luc Van Gool, Kooaba
Sept. 1, 2011 · 6:12 p.m.
136 views