Merken
Closing lecture with Anette (Peko) Hosoi
Automatisierte Medienanalyse
Diese automatischen Videoanalysen setzt das TIBAVPortal ein:
Szenenerkennung — Shot Boundary Detection segmentiert das Video anhand von Bildmerkmalen. Ein daraus erzeugtes visuelles Inhaltsverzeichnis gibt einen schnellen Überblick über den Inhalt des Videos und bietet einen zielgenauen Zugriff.
Texterkennung – Intelligent Character Recognition erfasst, indexiert und macht geschriebene Sprache (zum Beispiel Text auf Folien) durchsuchbar.
Spracherkennung – Speech to Text notiert die gesprochene Sprache im Video in Form eines Transkripts, das durchsuchbar ist.
Bilderkennung – Visual Concept Detection indexiert das Bewegtbild mit fachspezifischen und fächerübergreifenden visuellen Konzepten (zum Beispiel Landschaft, Fassadendetail, technische Zeichnung, Computeranimation oder Vorlesung).
Verschlagwortung – Named Entity Recognition beschreibt die einzelnen Videosegmente mit semantisch verknüpften Sachbegriffen. Synonyme oder Unterbegriffe von eingegebenen Suchbegriffen können dadurch automatisch mitgesucht werden, was die Treffermenge erweitert.
Erkannte Entitäten
Sprachtranskript
00:08
OK with that was the last session of the conference it's my great pleasure and honor to introduce Beckham was always has plenary speaker Sobek Ogata bachelor in physics Princeton and then went off to University of Chicago where she got the was clear :colon offered Dr. Paul in 1997 on the subject of suspension she Xiwen then went on to become postoperative corrupted in New York must apartment than later she was professor at a Harlem at the college and since 2002 choose professor at MIT commission agreed during and since the 2000 and also mathematics so you see she had been 3 different departments physics engineering and mathematics that was famous for work on free flow swimming complex roots nonNewtonian flow and robotics she has lots of collaboration with industry so hours and a stock today was designed as an categories and tackle clearly is a role model for the category of the of pasta impose fundamentals and applications so bad was famous for aspiring black shows that and also for aspiring teaching she got many teaching awards and she won various awards at the gallery of promotions at CBS division of food summit meeting and she is also famous for good taste of problems and 1 of her rare recent activities which I would like to mention as a program which he set up at MIT on sports technology and education at MIT briefly called scene today she will speak but on the subject were brought from biology To robotics but forces used thank you and thank them for that fabulous introduction and an end thank you so much to the organizers for giving me this opportunity in putting together this wonderful meeting and it's been really inspiring and exciting to see all the work that's going on in the Dutch physics community so I've really enjoyed the past couple days and all of a sudden I didn't tell you about biology robotics and an ominous start by pointing out what might be a controversial "quotation mark that appeared in nature in 2005 so disappeared as "quotation mark appeared in an essay by William Sutherland and says and their increasing calls for biology to be predicted optimization is the only approach biology has for making predictions from 1st principles and the wider adoption of the ideas right across biology should read ample rewards of it so we can argue about the use of the word only in that sentence and and I'm in fact that as an exercise for the audience so on your train ride home you can debate with your friends whether you can think of other ways for biology be predictive other than optimization and instead I'm focus on this last sentence and tell you about some examples of the rewards can be had when you apply the tools of optimization the biological systems and end in my group these are these problems sort of fall into 2 categories and 1 is to apply optimization to understand that at the start of the fundamentals behind the structures and strategies that we observe in biological systems so this is really just fundamental science questions and and 2nd is to take that understanding and then apply that to advanced engineering design and that's going to be the robotics portal all talk about so I mean give examples in both of these categories throughout the stock at Siemens start with I'm rationalizing biological structures and strategies and in the context of small swimmers OK so that's so small swimmers as you well know small is a relative term so I have to tell you small relative to what and when you're talking about swimming and typically when you look at the determine whether something is small it is the Reynolds numbers and so let me remind you just that the Reynolds number is dimension was number that quantifies the and the relative a importance of inertial and viscous effects and defined as you see appear so the density of the flu mu is the viscosity of the fluid you is a typical swimming velocity the swimmer and Ellissa characteristics 5 from that and so that the nice thing about biology is that these swimmers tend to swim in water or they tend to swim in fluids that have material properties that are very close to that of water it which means that you really can't play With the density or the discussing those are pretty much fixed so the only way to change the Reynolds number to change the size of the organism that so let me give you some numbers so you can calibrate yourself so you know what small means in this context and so a person swimming said to take a person and throw them into a swimming pool and a tennis women Reynolds number about 10 to the 5 and now as we go down in scale and if you go down something about the size of the duck that serve something like a thousand and you go even smaller you go down to the size of something like an ANC and an onto the Reynolds number of about 1 so this is somewhat surprising because in this context means answer not yet small right answer right right at the crossover where viscous an inertial forces are about the same so we have to go even smaller and so smaller goes around the corner of my life here as so over here and a pair museum would go down the size of the museum that also has a Reynolds number of about 1 the smaller than answer but they're better swimmers so the velocity goes up and then we go through there's a few things here that are not motel so they don't swim so we can skip get those the next limited on the charts about halfway up it's the bacterium which is about 1 micron inside and Reynolds number for bacteria is about at 10 to the minus force our finally done something that we can call small so this is something that is really a low Reynolds numbers swimmer and then again just to calibrate you this is sort of roughly the smallest thing that you can see with a light microscope roughly speaking it so as to the types of swimmers and be interested in are the ones the sort of between the pair Museum in the bacterium on this chart here so these are going to be singlecelled organisms or they're going to be bacteria and as you can imagine if you if you take thing I drop a pond water and put it in microscope whenever you see swimming around in there sort of fits into this category OK so now there's there's and there is another important thing to take away from this chart which is that if you think about sort of odd intuitive way we understand swimming we do all of this at this scale over here right attended the 5 writes all the intuition developed about swimming you've developed a very high Reynolds number and whatever he learned over there does not necessarily carry over to this part of the spectrum and I'm gonna show you an example of something that had to illustrate how are how intuition can fail there so this is a very famous movie and it was done by G8 Taylor said Jay Taylor is a giant in the field of fluid mechanics and he made this movie as part of the National Committee for fluid mechanics and yes in the 60's there was a national committee for fluid mechanics films in the U.S. state and an end the experiment Taylor does have the following have so he takes 2 concentric cylinders cylinder for my 2 concentric cylinders and the gap he puts viscous fluid and then he puts a drop of dye that red doctors to represent died a fluid and he's going slowly turned the inner cylinder to mix the diet that so here's the video so there's the apparatus it's very simple apparatus and so here is
07:47
concerning the die and those of you made it from you might be able to see the end of the year inner cylinders actually somewhere about here so it's a very narrow gap died fills the gap about halfway something like that puts the drop in
07:59
and now he's going to store it so will take the handle on the part of those in the surrounding the consumer drop is now getting mixed then as he does this thing he turns it 4 times or something that's twice around so 3 so far nothing exciting story exactly what you'd expect peso and enough good measures he's going to turn it the other way to get everything will next so he turns it this way they conceived sir the vague you can see the guys nicely mixed in there and so is in the background think twice around this way and maybe 1 more time around and it it is now a mixed case so and this is real there's no fakery here cake and and if you look at the equations of motion it's obvious that this has happened in the reason this happens is that aren't so you look at the equations of motion the balance inertia and viscosity and whatever other physical effect having there if you take inertia the equations time no longer appears explicitly in the equation so you have some inherent time symmetry in the problem so everything has to go back if you take inertia out everything has to go back to where it started and this is a big problem if you were a Chinese swimmer right because it means that there are certain things you cannot do if you want to generate some kind of net translation that so and so I'm not going to talk about that there are many many papers are written on that if you're interested in that Limoges give you this reference so this is a paper by Purcell called life at low Reynolds number which appeared in Scientific American and and this was that this is actually a transcript of a lecture that he gave at Harvard and it's a great introduction to the subject so he yeah it's like 4 pages long it's very conversational and high I highly recommend it OK so analytical backed up to our swimmers so our swimmers and before I talk about the physics I have to tell you a little bit about biology and I am not a biologist so all the while genitalia was burned in the context of of this problem I just have to slides on biology at the 1st woman show you is a fluid dynamics is view of biology so this is what biology looks like to a fluid dynamics I think it's immensely and it's a terrifying that yet so this is also I'm very famous image so this was done by James like was also another of great fluid analysis and so this image came from a newspaper in Siam review this is also another transcript of a lecture that he gave on flood alert hydrodynamics and there are many things I love about this picture 1st is that it was done in the days before PowerPoint obviously and then the 2nd 1 is that he is organized everything here 1st and foremost and not by any biological principles but by how things swim so if you look at everything in the center circle here so everything in the center circle has sort of the same morphology which means they the head which for this audience I can approximate as sphere and an end tale where n is a small number and that small numbers generally 1 2 or 3 OK so everything and everything in circle basically looks like this guy over here said that the 2nd the 2nd distinction driving here and is that he's put periodic excels at the top you carry out cells of the body and if you think back to your high school biology you might remember that prokaryotic cells are bacteria and roughly speaking you periodic cells are everything else so its people its grasshoppers it's penguins it's algae everything else states in here because it just bacteria appear eukaryotic cells and your head and acid this halfcircle at the top delineated the boundary between those 2 sides colored outlined in green and about it happens to be important in the context of the stock because the structure and the mechanics of the tail is very different between pro periodic new periodic cells so the cells under talk about in stock on the bottom half the eukaryotic cells and so the below the green line and did the red circle OK so the 2nd biology sliding to show you that I have to tell you a little bit about the structure of the tail is so here this is what the tail looks like any you carry on excel OK so the tale and
12:11
it has something that's called and 9 plus 2 microtubule structure so it basically if you look at the top diagram here there are out there microtubules that run along the length of the tail and there are 9 years ago around the outside and then to in the middle and 1 of the remarkable things about the structure is that it is constant across all species so if you look at the cilia in your lungs if you look at the full Geller on swimming algae if you look at the sperm on zebras or were Honda whatever your favorite almost all of them have exactly the structure tonight was to market structure which is remarkably consistent for for biology can so that's the 1st thing you know and this has 2 important consequences for this talk the 1st is that because the structure is the same the diameter of all the tapes tales of roughly the same across all species all all species so and the diameter is roughly something like 300 mm but that there is a little bit because you know is biology but it's very close to 300 meters per 2nd 2nd and these organisms on the way they moved the tail and if they have these microtubules and they can slide them relative to 1 another like this so imagine you microtubules at 1 end and then you slide them relative to to 1 another further down the further down the tale that mean you're going to induce a local bending moment and the tail will change shape so the organism can control the shape of the tale as a function of time they sell now that we that brings us to our 1st optimization questioning optimization question is located yes I can all things so the opposition question is yes I can't control what the shape of my tail as a function of time what shape should I choose so that's this question here can you predict the kinematics observed microorganisms orca organisms from purely hydrodynamic consideration that OK so so this I'm not sure some work that was done by my my migrating Donald 10 who is now at Delft and so did put together a beautiful model and I'm not going to go into the details of the model but basically we have and we have the hydrodynamics of the swimmer and their references here you could look up and do it and and the question we want ask is is finding the optimal curvature as a function of distance along the tales what shape should my tail take take and and you notice that this is a nontrivial optimization problem because I'm not optimizing wanted to parameters I'm optimizing shapes as a function of functional optimization has to be carried out that we can do it it's complicated but you can do it and I am and what comes out of this and this is actually a remarkably and again of the very remarkably robust as swimming strategy no matter what initial conditions we put in the always get something that looks like this 1 which looks very much like what you see in biological systems so you always get a traveling wave and his traveling wave is not a sign waved its regions of locally high curvature that are connected by sort of straight segments and so qualitatively it looks like what you see in biological systems and quantitatively does quite well as well as well so I had the computed optimal and ratio of wave amplitude wavelength is about 0 . 2 1 was measured and in this paper over here 0 . 2 0 and if I computer number periods for Taylor get while 1 . 2 3 at which measured as this 1 between 1 . 2 5 and 1 . 4 so it seems pretty good it seems pretty consistent and but in some ways this is not terribly surprising because if you only have 1 tale really what are you going to do mean that kind of the only thing you can do with rhetoric system so OK so we OK well let's go and think about something that's a little more complicated what's what's the next more complicated thing over 1 tale but it's going be too tales take plant tales so there are many many microorganisms that have 2 tales and probably the most famous 1 is Clemmie Dimona which is this appear but so climbing Morris a green algae biologist love 22 motors and there's lots and lots of data available for it and so we thought well since we have so much data should try to optimize the strokes received from modest and you immediately run into a problem because and it turns out that the algae that live very rich fulfilling lives and there are a lot of things that they're trying to do right because they can they need to eat and they need to escape predators and they need to find other Clemmie Dimona Snyder of the need to communicate with other Cunningham was ready so the question is what are you trying to optimizing the system it's not at all obvious with the cost function should be so we decided to take a guess and we thought ah I suppose suppose it is a I'm algae the most the 2 most important things to me are to eat and to keep from being eaten so let's suppose those are the 2 things that I'm trying optimized case of your mind also eat and has nutrient take or avoid being eaten so that's outrun press predators and you don't have to do those at the same time right so at some time you might be doing a stroke that will enhance your nutrient uptake and sometime you might be escaping and so on and so here's the answers you get so if you run Then the optimization for this outrun predators and you get the strokes so these are the 2 escaping strokes that with something like that and actually um what Daniel found was for local minima for escaping strokes but I'm just gonna show to hear because 1 of them is 1 of on part of the street to hear play the white cake and now if I if I run the optimization for enhancing nutrient uptake so this is the eating stroke and you get something that looks like that it's very nice and and you can make some
17:36
pretty movies so you can put on some this with the work history field looks like a few computer everything and so like this and there were 4 local optimal that came out of this and 1 that came out of this and those 5 strokes for the strokes that are observed in the living climbing toward the end of 4 and I have a very nice to be here from a Goldstein's grew out of a climate Morris and actually exhibiting 3 of the 4 OK so this
18:01
is a clever this is the climate Morris here it's stuck pipe that is currently doing the feeding stroke uninvited defeating stroke is the most common ones of it's just hanging planning once a detainee at the has announced it switched over to the escape stroke and then you can trigger despite by something called a shock response so you can chop them would like to get them to to go in 1 of his most recent now it's gone into the 2nd escape also this is now and the antisymmetric escape mode and then in a minute is going to go back and now it's back to its hanging out lowered his feet that so and so in fact you can now so it's so it's so so that the opposition routines actually do quite a good job finding the kinematics you actually see alive organisms so I thought OK that's great let's move on to the next question so what's the next question I ask is that OK we can get can kinematics quite nicely and what about morphology can we predict the shapes the morphology observed microorganisms from Julie Hagerty and consideration and destiny is actually a more interesting question in some way because so imagine the following imagine someone takes you and throws you into a swimming pool and you're going to flail around and you're gonna selected kinematics until you find something that will allow you to swim right so you can sort of sample a lot of the phase space to figure out what is the optimal stroke the morphology cannot change you cannot go into a swimming pool and say gosh I really like a 3rd arm you and it's not going to happen ahead so morphology is something that is fixed for you and can only change the evolution right so this question actually gets back to the roots of what is the role of physics and hydrodynamics in evolution OK so and so here's the question we want so so for this problem we decided to focus on sperm and the reason we decided to focus on sperm is that they have a very welldefined objective function right they have a packet of genetic material and they have to move it from 1 place to another and that's it they don't have to eat they don't talk to the sperm there have been doing that they do that is purely transport Quebec so we got him a welldefined objective function and so here's the question we want if you have a packet of genetic material which is in the head case it's packaged in the head here and how big of a motor should I put on that day in the motor is the tale of how long the Telstra on and it's if you think about it and that there should be an optimal size because if your tail is really small then you obviously have going to go anywhere and if your tail is really really long then you're expending a lot of energy moving the tail and relatively little moving ahead so there should be some optimal optimal telling between us and the it's so that so here's the here's the plot that Daniel they said this is basically swimming efficiency and since there's only 2 length scales the problem the length of the tale of the size of the head at this should really just depend on the ratio of the 2 which it does and I should also say that there is a lot of computation the goes into this plot because for each 1 of these points so let's think this point here so this point is kick ahead size head size italicized size now optimization for those 2 guys to find the best possible kinematics and then recorded swimming efficiency for those kinematics right so these guys might not be swimming the same way these guys are and what we're allowing everybody to sort of put their best foot forward and swim as best as you possibly can and given that morphology picks itself is always it's not fair to fair competition again so everybody's doing the best they possibly can and and this is 1 of my favorite plots that assume has ever brought to me and the reason this is 1 of my favorite plots is because there's an answer and the answer it is 12 Bloomberg currently 12 again so that remember the question how large a detailed be well actually 12 times the size of that and you know that it does all these other details don't matter so then the next thing Donald as he went to buy a biology literature and and he looked up the morphology of over 400 different mammalian species over 400 different species and then he made a histogram so here's what you get from biological data it's toward all over it's great so OK so I was really excited and really excited about this I when I gave his talk in physics departments and engineering departments in biology department and 1 of the things I learned is that biologists and physicists see something very different in this plot so this is an engineer's they tend to look at this and say all this is wonderful Look we know we've learned something fundamental about the way things swim this is great and biologist look at this and they say this is completely boring right the interesting part of this plot is the outliers because the outliers are either I their suboptimal which is kind of like the war they were subject to different evolutionary pressures and constraints which is now an interesting biology question but kept silent OK that sounds good right so I thought OK we might look at the outlier so let's take a look at the outliers based on a show you what 1 of these outliers arcade before and we've had to give you an exerciser before I tell you what the answer is I want you think of the weirdest mammal you can get so you have 10 seconds to think of the weirdest mammal that you can think of no better tell you the answer they self I give 100 per cent credit to anybody who gets the band he couldn't I
23:32
have no idea what a bandit who was before I die before I did this day now turns out that the baby who is a marsupial so I also gives partial credit to anybody who gets marsupials replied to right the filet books with something family get so no more I wide the band could sit out here they sold to answer that I have to go back to a slide show you earlier so the slides actually earlier as I said that I said there's this amazing structure this mindless to microtubule structure which has the same diameter across all species and when I should have said is there's this
24:05
amazing structure which has the same diameter across all species except for maybe said because there's always an exception in biology get so now wiser than couldn't acceptance and so we figure this out because there is in fact a paper that was written in 1958 called the includes Berman is all in an electron microscope study of the tale is so this is so this is a biologist measured everything is OK so stable and here's the electron micrograph and any fact it's still the mindless to microtubule structure right here mindless to microtubule structure it's sitting here but it's in this big sheet cakes so which means that the radius of the tail is now something like an order of magnitude larger than what you see in the normal in the normal tales and this is a problem because this is his you know that if you're radius goes up your bending energy goes up enormously goes up something like lot the 4th right so you know there's an enormous bending cost now associated with is that we did not have an original population and in fact when you put bending cost and at what it does is it moves this peak over this way to where they said and and all of these guys down here are animals that had the sort of funny factory or whatever so that's those those are those outliers cancer because I think we kind of understand and but there's another set about large which we do not understand so I'm just going to tell you what the what the puzzle but I don't have an answer for it so let's think about what the answer might be so so if you look at the history so the colors on its histogram correspond to different waters so waters meaning like carnival words or rodents or primates etc. etc. and if you take any 1 of these any 1 of these orders that has a sufficient amount of data in it they're all sort of centered around 12 just like you would think that with the exception of the yellow guys over here In the old guys obviously are not centered around 12 the center on something like 6 right and so their way off and effective you look this way efficiency is not you know it's at least 50 per cent less of no more than a year of not more than 50 so and so I don't know why the is statistically significant we have a lot of a lot species in there but I can tell you what they are so the ID animals there are in this yellow here or something called even towed on Gillette's OK and he even told undulates or who lived in animals so the things like pigs and goats and cows and sheep and and and that and again I like is that everyone is talking biology department people don't know why these are different but I think there's a nice and there's an interesting open problem here to understand why that water is fundamentally different so OK so so that's what I have to say about small swimmers and ownership years talk talk a little bit about robots OK so and so
26:49
various announcements on mobile inspired design so I inspired design and so this is not a new idea people have thought about inspired designed for many many years and so here's maybe 1 of the most famous examples and you guys might recognize this so this is the bird that inspired the invention of Velcro in
27:08
1941 so this spend an afternoon picking these out of his dogs for about while this is a really good adhesive and indeed Velcro K amp but of course quite inspired design actually has far more failures than it has successes and if you go on YouTube you'll see many many views on flapping flight which too as you can tell it's completely ineffective OK so the
27:30
question you have to ask his OK if I'm going to do while inspired design and what I do To make history I noticed yeah you that I have relied on that last part of the movie because he had to be really sure it was going to work version of the product .period gets itself so the question to ask yourself is how you how do you ensure that you are in the Velcro billionmaking Velcro and then you're not trying to make this project so and so we have a couple of basic rules of thumb that we use it in in my
28:06
group and so the 1st is that you should choose a simple organisms so preferably something with a very primitive nervous and central nervous system and even more preferably nobrainer break because I don't have to build the brain I wanna do something with the challenges and solutions lie in the mechanics rather than in the neurological controls rights so I'm looking for organisms that have mechanical solutions to the to the challenges the 2nd you won't find an organism that is in some ways orders of magnitude better than current than the existing engineering technology because we already have something that does what we wanted to do then you why would you go to this complicated biological mechanism that soandso in some way it has to be better and that better better is a very I'm deliberately left that is a vague term because it could be inefficiency or robustness or simplicity there are lots of ways things can be better but somehow it has to be better than what we already have on the 3rd and probably most important point on this is that the goal here is not to mimic biology the goal is to understand the underlying physics of the biological solution and then adopted in engineering design OK take so and so we've done we've applied this many many different organisms and entered on the talk about 1 which is my my mostrequested organism which is at the snails OK so snails snails are incredible so snails if you take a snail is put on the grounds it'll crawl across the ground until it's a wall in the middle crossstrait up that wall namely the ceiling will flip over crossstrait across the ceiling it doesn't matter if this is grafted dirt or sand pocket whenever it crawls over any kind of substrate you wanted any angle you want so wouldn't it be wonderful if we could build a robot that had the same kind of versatility as snail that so that was that was oddballs bolstered let's build robotics they'll get now snails the way they crawl at the basic rate of thinfilm of fluid that sits between the foot and the and the substrate case and the only way that the foot interactive substrate is by generating stresses in this felt it never directly touches the substrate and the way it generates the stresses and is by driving waves along the bottom of the foot so this an international will be of a snail crawling up a plate of glass and so you can see those waves and if you eat if you find if you catch a snail you can put on a plate of glass and available to see this and soulful you can see waves moving up history of waves that are moving up along the bottom of the 1st case and and if you stare at this for a while and stared at many of these images for for a very long time and you start to think whether something really bizarre looking about this I mean other than the obvious fact that the bond sale but but there's something very bizarre looking in the recent it's bizarre is that the waves are going in the wrong direction right if this thing was going to pursue pushing off the ground away should be going backwards as thing goes forward so already there's sort of a puzzle here OK so and so this problem was actually brought to my attention by my graduate student Brian Chan and so Brian chance he got interested in sales he came to said work on stale project as a volcano I know much about snails wanted you go and do a literature search come back in 2 weeks tell me there's something interesting for us to research and then we can decide whether or not to look at sales the right so he comes back 2 weeks later and then I said OK so and so on what would you would you find out about sales Bryant and he said Well I solved it and at I don't even know what the question was was going to wanted to solve a sizable Mark Bryant wanting you solve it wouldn't you do and he said I built a robot and in case you may have to understand this is I was the faculty at MIT at time Brian was the 1st PhD student I've ever supervised and at that time I had not appreciated that if you give an MIT student at proper problem with no would be sufficient constrains what they do if they build a robot which could get some say so bright and build a robot that could end at any track that interestingly build a robot was exactly the right thing to do in this case association Brian's robot as this Robles smell so far so here in the eye but the 3rd is to use a rigid pieces and the speed that they're attached to each region pieces cake and and and you activate the robot in the following way so there's a time series ago is a time series of the top here and so 1st you take that the rearmost pattern sliding forward and then the next 1 and 545 courtside for forward and you get is wave of compression the travels along the bottom of the of the stale cake all and I should like to actually mentioned on the movies saw free on the previous slide with still crawling up the glass the waves that you saw we're not out of playing weight so that they're not waves that do this they're waves of compression that to this day and you can and you can measure that using interferometry or other kinds of tricks right so that so that the thickness of the film actually changes very little it's the wager saying something like that OK OK so this is what Bryant sailboats on and we can now do the backoftheenvelope calculation to figure out whether or not the single crop is so um 7 years the model that
33:09
I'm not going to do just a force balance and on this last trump over here so the forces that are acting on that show are basically the forces that are exerted by the muscle or the motor or whatever you wanna say that's pulling this thing forward and that's resisted by the Forsa yet from the shear stress in the fluid over here it so I am now that means so that means this force over here the muscle force has to be balanced by the forced mission by the sheer stress and fluid gave this is this the shear stress is well known and this is something called Colette flow in a small gap here and we know that the shear stress is inversely proportional to the gap that I am nearly proportional to the velocity which is the top plates and to the viscosity of this fluid that and you do the same force balance on this other child and and then you just say that the velocity of the center of massive sale the average velocity of all the fees and and then you come up with an answer and it takes 3 lines or something you can do it on a napkin and and you get the following so the velocity of the snail looks like this and there are a couple of interesting things about this calculation so as a 1st assessment prefectures but the the most important part is that is this guy over here so this says that the snail velocity is proportional to 1 over the viscosity that that had once these minus 1 over the viscosity but the rest of this crisis it now so what does that mean that means if I have a fluid with a constant viscosity this thing cannot move so just at there and of this they said they it's basically amuses treadmill in something jogging in place said so does that it now however as soon as I have a kind of nonlinear reading discussing if anything like it the viscosity in some way depends on local shear stress to the local charade or output put my putting anything that that deviates from constant the will crawl cake and furthermore the type of nonlinear nonlinear put determines the direction it goes right because there's there's a minus sign here right so I can put the sign on this by flipping it around so if I have something called a sheer sickening fluent in which my viscosity increases as the shear stress increases then the wave goes in the same direction that the snails crawling if I something called the sheer influence so the viscosity decreases as the the shear stress goes up the way it goes in the opposite direction to to with still strong and fact you go and look at snails and there they exhibit both types of behavior land snails and which have very sheer thickening sheer thickening and mucus have weighed the going the same direction sales falling and we know that land sales mucus has this property because there's a Ph.D. thesis by Mark that was published in Nature in 1980 on the real real logical properties of snail slime marketing when I talk to me had no idea anybody whatever site that paper but it's turned out to be incredibly important OK so now the last nevertheless thing will show you is of course to the rubble still climbing at a snail so it's very slow but and understand this is a movie of Robles still Ponte climbing up a glass plate and so we had no and we we had no snail slime in the lab so we use whatever we had his like on the previous slide we can use any kind of fluid that has it's nice not nonNewtonian property that it
36:19
climb tiedown unsealing it's actually easier to go on feelings than to go up walls right because up walls you're fighting gravity both in this direction and grandees trample you off the wall but on feelings and that is just trying to play ball but it's not impeding your progress made so that this is actually easier thing and it's running now and ,comma which is there and it's a commercial product that they put in here just so it's clear that has yield OK so an endowment would just them the most most important financing which is the people who actually did the work and so at 1st Daniel Tamm who did all the work on the small swimmers and is currently an assistant professor and in in Delft and he's 1 of the locals know Bryant Shanahan and is currently stands at MIT said the guy that built the robot in radio and I wanted throw out there even I didn't I didn't talk very much about the reality of snail slime he did a tremendous about on the reality is still flying which helped us to optimize the snails hours crawling and then I talked to many many many many biologists who taught me a lot of biology probably the most important limiter at Harvard and Suarez at Cornell and would like to thank you for your attention further but flows were recent talk so the fall somewhere like to take some questions the questions there was some question over there do we have my Christian Schulte we all know what we need yes that's a great question so that the the question was in in our simulations all of the motions of the tail seemed to be in plain where is the real world is obviously 3 days so that the so things can in fact moving in another direction so absolutely and they're so so typically speaking in effect when we go back to this 1 slide so if you look at the at the microtubule arrangement the back here and you notice that the symmetry is actually broken in this tale of I'm Pakistani OK never mind that get so remember there 2 there 2 microtubules in the center which break the symmetry of tales of this tale actually prefers to bend in a plane now having said that there are a lot of organisms that actually do that actress when using I've been in detail in a spiral areas so you see here right here the symmetries broken so tends to like the band along either axis so most of the swimmers of the data that we looked at they actually do bend detailing in a plane and which is not again it's biology so there are always exceptions and I think the ones that actually that actually is a spiral are very efficient swimmers just yet we need so requested so questions has Robles still had any practical applications and the answer is yes so that research was actually funded by Schloemer and Schlumberger was interested in developing crawlers for the downhole drilling environments so it ended in the drink it when you drill a hole for a for all oil recovery and you basically you send down a slurry and you you know you you sort of grind of whatever's down and then you have to pump everything back out which means that you have a to that is full of something called drilling mud came drilling muds a very nonNewtonian fluid it has hasn't it has a yield stress it's sticky stuff and the problem is that if you send a robot down and gets stuck you're done you have to drill another hole there's nothing you can do to get it out rights of what they wanted because they want a robot that would not get stuck and this Robles still is designed to crawl on exactly that kind of fluid so the stickier better writers closer itself was so and so after we published this week passed that on to petitioner's age were now developing that those ideas into crawlers for their for their downhome environment With the was walking working for toward the theft
00:00
Greiffinger
Schaft <Waffe>
Anrufbeantworter
Spiralgalaxie
Mikroskop
Messung
Magnetisches Dipolmoment
Färber
Gleichstrom
Dünne Schicht
Geokorona
Satz <Drucktechnik>
Bohrmaschine
Minute
Leitungstheorie
Computeranimation
Schlauchkupplung
Radioaktive Strahlung
Schleifen
Biegen
Scheinbare Helligkeit
Bohrspülung
Trägheit
Amplitude
Nachmittag
Abwrackwerft
Stromschiene
Räderuhr
Konzentrator <Nachrichtentechnik>
Trägheitskraft
Elektronisches Stabilitätsprogramm
Interferometrie
Längenmessung
Grün
Mikrofilm
Gigant <Bagger>
Hydraulikleitung
Gelb
Klettverschluss
Rootsgebläse
Ausreißer <Messtechnik>
Übungsmunition
Angeregtes Atom
Werkzeug
Konfektionsgröße
Motor
Satzspiegel
Schwimmer <Technik>
Segelboot
Jahr
Reihenschlussmaschine
ETCS
Hohlzylinder
Natürliche Radioaktivität
Kristallgitter
Walken <Textilveredelung>
Glas
Fliegen
Gewicht
Niederspannungsnetz
Feinstblech
Masse <Physik>
Höhentief
Newtonsche Axiome
Maßstab <Messtechnik>
Schnittmuster
Hobel
Begrenzerschaltung
Digitalschaltung
Umlaufzeit
Kopfstütze
Videokassette
Lautsprecher
Herbst
SpeckleInterferometrie
Energielücke
Stunde
Gasdichte
Tag
Ausschuss <Technik>
Druckkraft
Schwarzes Loch
Substrat <Mikroelektronik>
Gleichstrom
Nivellierlatte
Hintergrundstrahlung
Fluidität
Fluidität
Sprechfunkgerät
Material
Ersatzteil
Undulator
Dissoziation
Zwangsbedingung
Erder
Potenzialausgleich
Schwimmer <Technik>
Zelle <Mikroelektronik>
Kugelblitz
Elektronenmikroskopie
Konfektionsgröße
Rutsche
Material
CocktailpartyEffekt
Cross over <Teilchenoptik>
Optisches Spektrum
Schwächung
Sutherland, William
Schwache Lokalisation
Woche
Regelstrecke
Passung
Zelle <Mikroelektronik>
Fahrgeschwindigkeit
Kristallgitter
Entfernung
Raumfahrt
Drehmasse
Druckkraft
Wellenlänge
Kaltumformen
Pager
Rollsteig
Unwucht
Makroklima
Stoßdämpfer
Gleiskette
Gruppenlaufzeit
Handy
Regentropfen
Plattieren
Druckfeld
Öffentliches Verkehrsmittel
Lineal
Dünne Schicht
Anstellwinkel
Photographische Platte
Regelstrecke
Regler
Leitrad
Besprechung/Interview
Erdefunkstelle
Wellenlänge
Feuerwehrfahrzeug
Wegebahn
Elektrotechniker
Förderleistung
Druckmaschine
Druckluftanlage
Fuß <Maßeinheit>
Zylinderkopf
KraftWärmeKopplung
Brennpunkt <Optik>
Entfernung
Waffentechnik
Gruppenlaufzeit
Gleitsichtglas
Videotechnik
Plasmaschicht <Zweidimensionales Plasma>
Nassdampfturbine
Grundfrequenz
Messung
Zylinderkopf
Buntheit
Leitrad
Mechanikerin
Masse <Physik>
Papier
Metadaten
Formale Metadaten
Titel  Closing lecture with Anette (Peko) Hosoi 
Untertitel  From biology to robotics 
Serientitel  Physics@FOM Veldhoven 2013 
Autor 
Hosoi, Anette (Peko)

Lizenz 
CCNamensnennung  keine kommerzielle Nutzung  keine Bearbeitung 3.0 Deutschland: Sie dürfen das Werk bzw. den Inhalt in unveränderter Form zu jedem legalen und nichtkommerziellen Zweck nutzen, vervielfältigen, verbreiten und öffentlich zugänglich machen, sofern Sie den Namen des Autors/Rechteinhabers in der von ihm festgelegten Weise nennen. 
DOI  10.5446/18012 
Herausgeber  OpenWebcast.nl, Foundation for Fundamental Research on Matter (FOM) 
Erscheinungsjahr  2013 
Sprache  Englisch 
Produzent 
OpenWebcast.nl

Inhaltliche Metadaten
Fachgebiet  Physik 
Abstract  Professor Anette (Peko) Hosoi has been an Associate Professor of Mechanical Engineering at MIT (Cambridge, US) since 2006. She received her PhD from the University of Chicago in 1997 and first came to MIT as an Applied Mathematics Instructor from 1998 until 2000. She joined the Faculty of Mechanical Engineering in 2002. Hosoi is a specialist in free surface flows, surface tension, and the fluid dynamics of complex fluids. From 2004 until 2006 she was the Doherty Professor in Ocean Utilization. She has won numerous teaching awards, including the Ruth and Joel S. Spira Award for Distinguished Teaching and the Junior Bose Award for Excellence in Teaching. In 2010 MIT selected her to be a MacVicar Fellow. 
Schlagwörter 
biology robotics 