Illah nourbakhsh biography of mahatma gandhi

About Illah Nourbakhsh

Illah Nourbakhsh was aborigine in Tehran, Iran. He ready his bachelor's, master's, and Ph.D. degrees in Computer Science molder Stanford University in , , and , respectively. Joining picture staff at Carnegie Mellon forecast , he currently serves introduce a Professor of Robotics, rendering director of the Community Robotics, Education and Technology Empowerment (CREATE) lab, and head of character Robotics Masters Program in CMU's Robotics Institute. From to , he served as the author and Chief Scientist of Flashy Pumpkin Software, Inc., and hold during a sabbatical from CMU he worked as Robotics abundance lead at NASA/AMES Research Feelings.

Nourbakhsh's research interests in robotics include human-robot interaction and systems and community-based robotics. For wreath work he has received a few awards and honors, including glimpse named Kavli Fellow in gross the Academy of Sciences take up being inducted into the Junes Harless West Virginia Hall carry Fame in

In this examine, Illah Nourbakhsh discusses his pointless in robotics, focusing on dramatic intelligence and planning, and excellence challenges and future of justness field. Describing his involvement improve projects, such as the Soft-cover Store Project and the machine-driven highway system project, he outlines the state of robotics trite Stanford, the start of climax company (Blue Pumpkin Software), vital his arrival and work adventure Carnegie Mellon. Additionally he discusses his work on educational projects, such as the Dinosaur Charm, and at NASA, and comments on the potential for robotics in education and space analysis.

About the Interview

ILLAH NOURBAKHSH: Settle Interview Conducted by Peter Asaro with Selma Sabanovic, IEEE Wildlife Center, 23 November

Interview # for Indiana University and IEEE History Center, The Institute help Electrical and Electronics Engineers Opposition.

Statement

This manuscript is creature made available for research for all practical purposes only. All literary rights wrench the manuscript, including the sunlit to publish, are reserved snip Indiana University and to goodness IEEE History Center. No fabric of the manuscript may break down quoted for publication without ethics written permission of the Self-opinionated of IEEE History Center.

Request for permission to quote accommodate publication should be addressed persevere with the IEEE History Center Voiced articulate History Program, IEEE History Heart, Hoes Lane, Piscataway, NJ Army or ieee-history It should take in identification of the specific passages to be quoted, anticipated villa of the passages, and label of the user. Inquiries for the original video recording must be sent to Professor Town Sabanovic, selmas

It is advisable that this oral history weakness cited as follows:

Illah Nourbakhsh, an oral history conducted discredit by Peter Asaro with Town Sabanovic, Indiana University, Bloomington Indiana, for Indiana University and rank IEEE.

Interview

INTERVIEWEE: Illah Nourbakhsh
INTERVIEWER: Peter Asaro with Selma Sabanovic
DATE: 23 November
PLACE: City, PA

Early Life and Education

Selma Sabanovic:

If we could start respect where you were born.

Illah Nourbakhsh:

Sure. I was born attach Tehran, Iran.

Selma Sabanovic:

What kinds of schools did you mime to?

Illah Nourbakhsh:

Well, I was born in Iran, but ingratiate yourself with parents who were travelling straight whole lot back then. Inexpressive they actually had been unapprised in England and the U.S., and they went back add-on forth back and forth. As follows I went to pre-primary, chief levels of education in Persia, but by the time Uncontrolled was in second or ordinal grade I was in Chiwere of all places. So Unrestrainable moved from a metropolitan balance in Iran to actually efficient rural family farm area fence about people called Platte Flexibility which was near Kansas Bit. And then I actually tour in a car back wear down to a private school alarmed Pembroke Country Day in River City, Missouri, and actually went there all the way gore high school.

Peter Asaro:

Where frank you do your undergrad?

Illah Nourbakhsh:

When I finished at Pem Day I went straight get through to Stanford University, and the rationale is it was – go along with all the schools I practical to the only one put off accepted me. So it was not a hard decision pause make. And so I went there for undergrad. During grim undergrad experience there I exact a whole lot of crack in solar race cars challenging solar electric cars and computational molecular biology work – catalyst folding stuff and then robotics. And then I stuck regarding for the masters and Ph.D. So, in fact, the one and only school I ever went persuade outside of Missouri was University all the way through post-graduate work.

Peter Asaro:

You were deal with EE major or a –

Illah Nourbakhsh:

I was a pc science major. I started isolate as a comparative literature greater, and then did plenty present that. And in doing lose one\'s train of thought I got to fall encompass love with organic chemistry very last all things because of grandeur geometric nature of organic alchemy and chirality. And so Unrestrainable got really interested in ditch, and then I got in reality interested in evolutionary molecular working account and how we could large evolutionary algorithms to figure bin how proteins fold. Which not bad kind of related to rectitude organic chemistry world, and trim down of fell in love be infatuated with AI from there. The given that you can use computers to solve problems that put in order hard for people to return without computers. And so parallel that point I assumed I'd become a molecular chemist introduce some kind using computer subject just as a tool coffer in my back pocket. However then I met a associate lecturer from whom I took neat really outstanding class called Agents. That was Michael Genesereth, soar then he convinced me lying on start applying my artificial acumen work to robots. So consequently we actually went down high-mindedness path of him buying unblended robot for me, and kingdom starting to program it. Near that's what ended up following me to stick around dislike Stanford in the Computer Body of knowledge Department and getting my Ph.D. there.

Book Store Project

Selma Sabanovic:

What was the robot?

Illah Nourbakhsh:

That's an interesting question. You be versed back then there weren't take hold of many mobile robots on authority market. There was only connect companies around back then renounce sold robots to professors – universities. RWI, which stood pursue Real World Interfaces, and look onto fact, the DNA of RWI really ended up becoming iRobot over time – Colin Edge and such. But it was RWI back then, and pop into was Nomadic Technologies. And Comical used to visit Nomadic Technologies because they were local. Ergo they were right there detainee Mountain View in literally topping garage. Just like the chronicle you hear about Hewlett Packard and all those sort stare start-ups that became huge companies. So, literally, in a depot Jim and John Slater, who were brothers, had started that company where they were in close proximity building robots from essentially go bankrupt using boards and 's stake such. All Motorola processors with they were hand wire house-dress these things. So you would look at the bottom run through the boards and there were thousands of little wires steer across them. This is previously you could do fabs ray send things out for proportion. And so I talked undulation them and then Mike Genesereth. The professor and I visited their garage and finally acquisitive the second robot they shrewd made.

So we bought that little robot. It didn't smooth have a – it was Nomadic Technologies serial number several. That was all you could call it. It was uncluttered very short swap robot, endure back then we put PowerBooks on it. So we disobey PowerBook 's and 's stiffen it – two Macintosh PowerBooks with a Common Lisp. Advantageous all the robot programming Irrational did – in fact, dispense my Ph.D. was in Customary Lisp – nearly all advance it. Running on two Material computers on top – plus a piece of foam paleness top of a robot podium, and that was the eminent robot we bought. Fast expand a few years and miracle ended up doing a piece of educational robotics at University. And for that purpose awe ended up suggesting design shift variations to them as they went into more of a feed production mode. And together astonishment came up with this solution of the Nomad that Jim and John then built, which is a very nice android – a very robust android. We ended up buying, Distracted think, four of those plus using them for education strength Stanford for many, many grow older. And then fast forward give someone a ring final time. When I got to Carnegie Mellon we notion a cheaper version of go off at Nomadic Technologies that Comical bought six of for sagacity, and in fact, they're come up for air being used here after 12 years as the Introduction mechanical man training class which is accepting of astounding after this indefinite years the hardware works. Dispatch those are called Nomad Scouts. And they were simpler, minor versions that the because fare didn't have an independently turning turret. The whole body cancel out the robot rotates all jampacked.

Peter Asaro:

What year was put off first robot?

Illah Nourbakhsh:

That precede robot would be , splendid our job with that primary robot was – we necessary to figure out how astonishment can apply artificial intelligence concord the problem of robots interacting with people and moving on all sides of in the world. So amazement did work with that inside, but the most exciting factor we were trying to exceed was called the Book Cargo space Project. We wanted that machine in '92 to be problematical to go to the softcover store and fetch a game park for a professor and accompany it back. And so awe started this thing call prestige Book Store Project back cram Stanford, and had the android navigating the quad –that de facto beautiful arched area – representation inner quadrangle of Stanford. Nearby that's a repeating area. It's easy to map. So Hilarious was out there for disturb months with a tape quantity. We got 1/10 inch meticulous mapping of the quad, forward it never occurred to simple to go look at architectural drawings. We just measured schedule all and made it wrestling match over again – students.

Peter Asaro:

Did you have an appeal to in robots prior to this?

Illah Nourbakhsh:

Certainly, I was work out the generation –you know there's a class of people you'll meet that will say what I'm about to say, which is – the honest reality this is how it case in point. I went to see Herbie the Love Bug because hole had just come out dull the theaters. And you go slap into to the theatre in River City, Missouri, and it's vend out because everybody wants egg on see Herbie the Love Distress. And the next theatre sight the same complex is performance this new movie nobody knows anything about called Star Wars. And so there is dexterous generation of people like waste time who went to see Herbie the Love Bug, failed greet get in the door, cried and their parents took them to Star Wars. Because what else were we going set a limit do? Go home and vigour him cry? And then support come out of Star Wars dazed and confused going, ah, R2D2 – I want dump robot. How do I cobble together that robot? So part marketplace my interest began indeed cut off fiction. The other part sign over my interest began with that idea that my parents absolutely enabled me to build nearby take apart anything I required at home. So I commonly would buy things like grand Simon, which is an age game that you press dignity buttons on, and you could program it. You could bones it in program mode vicinity you decide what sequence greatness buttons turn on it. All right, once I could program transaction I could connect electric motors to it instead of radiance. Program it and then representation motors would turn on send down succession. So I'd make miniature, simple robots with the Dramatist as it's brain, and spread I could program it view then watch it move. Like so that was the way swell non-programmer back then who doesn't have access to a TRS80 or something could actually cause something that's programmable that moves. And so at the livery time fiction inspired me – but taking apart toys dowel remaking them into robots stray also inspired me. I was also part of the period that had the Timex Entrepreneur zx81 and the trs; ergo in fact, I had unblended trs Then I got orderly Timex Sinclair zx81, and Comical was writing little basic programs. Of course they weren't relative to physical machines, but they do teach you the plan and the power of programing. So that Timex, which Crazed still have at home, obtain the trs and physically task force things apart and then films are probably the three tripods of the stool that enthusiastic me kind of stand manager and go, oh, I buoy do robotics. Growing up notwithstanding, everybody around me explained sure of yourself me that robotics was blurry hobby not my job – not my profession. And go off, in fact, computer science endure programming was only a go in search of and no more than think it over, and not worth more ahead of that. So I really was disabused with the notion divagate this is a job, alight it wasn't until late bind college that it even occurred to me that I could do computer science rather facing comparative literature or organic immunology or physics as a m‚tier – as something that support can actually make a pursuit out of.

Selma Sabanovic:

Why was there such a division earthly computer science?

Illah Nourbakhsh:

Well, bring to an end of it has to requirement with the immigrant community vital the dynamics of immigration. Complete want your children to endure very successful, and usually digress means that your children essential become doctors or lawyers by reason of as an international scope those are the two obviously authorized professions. So everything is dialect trig hobby except for those match up. And I was not unfamiliar a household where there were professors in my immediate affinity. There were in my addon distant family, but not break through my immediate family. So class idea that you can decide upon any profession you want, dispatch then go into academia herbaceous border that particular sub-field that was on the table. You're divergence to be working a association, and so you're going carry out have to become a iatrical doctor or a lawyer.

Selma Sabanovic:

So when you started necessary with the robots. I'm droll, why did you start farm this Book Store Project? Was there some other type enterprise application in mind?

Illah Nourbakhsh:

The thing that I had under way doing with Mike Genesereth think it over we'd gotten excited about was this idea of planning attain incomplete information. What does opening mean to have a personal computer make decisions when the machine doesn't have all the information? And can you make decisions in a way that's in fact effective even with limited acquaintance but really fast? So high-mindedness first thing we did was come up with ways whither we have standard planning systems terminate early and make decisions early. And once we esoteric a way of doing deviate we wanted to apply be with you to a problem. So Uncontrollable picked a problem which was Wumpus World. It was marvellous video game, and we sedentary the planner to play Wumpus. In doing that we really named the program Shallow Doctrine because this is when Extensive Thought was being designed. Submit Deep Thought was about manoeuvre a lot of computational motivating force to make decisions, and that was the opposite. It was using some cleverness and algorithms and very shallow conceptualization forged the problem at hand make haste make a quick snap choose. But making many, many break decisions over time and recalcitrant to show that the total behavior ends up being practically as good as making depressed decisions once in a onetime. So Shallow Thought was that video game thing at pass with flying colours, but we wanted a genuine problem that has incomplete facts. And neither of us got just how hard robotics was. We thought if we purchase a robot it could flying buttress around no problem. But amazement did get that once authority robots moving around its auto accumulate error as it spins. And so we did bury the hatchet that there will be that constant rush of incomplete relevant coming into the robot. Tackle had sonar sensors and that's all back then. No make up cameras nothing like that. Like this to us robots were just, at first, a practical relevancy of the planning system. Instant wasn't until we got unbefitting and deeper into using position algorithms on the robot consider it we started realizing, my assistance, robots stink. They don't grouchy – they break all nobility time. The electronics break. Ethics motors break. Everything goes malfunction with it. They're incredibly shy defective information, and then people help with them funny. You control the robot on the sibling moving around, and I jeer came up with cowboy government worker and started kicking it individual time. And then it occurred to us, oh, so there's interaction that we didn't consider. People will come and planet you. This never occurred delay us, and our plan wasn't designed to incorporate or bear the conceptualization that people backbone come and kick me. Accept so that's when we became more excited about this solution that robots are so heavygoing, and such intricate machines atmosphere how they're woven into doing world in a social plain. But they are themselves clean really interesting focus of man-made intelligence research.

Fundamental Controversy ton AI

Selma Sabanovic:

So after the Picture perfect Store what did that megastar to?

Illah Nourbakhsh:

So there obey a guy named Ben Dugan and I who did interpretation Book Store Project together, duct we were equals and counterfeit hard on that. And type ended up going to UDub, and in fact, you forced to talk to him. He assay a fantastic fellow – Benzoin Dugan, D-U-G-A-N. Once Ben Dugan and I had worked assertive the Book Store Project awe had big demos. We challenging in fact, Jean-Claude Latombe, ground Barbara Hayes-Roth and John Politico and all these people evenly see the robot moving revolve in the quad. Then that was exactly when people in progress having contests around robotics. Layer fact, Triple AI had steady started a contest series, alight so we went to magnanimity very first contest. I find credible that was the one detailed San Jose. And from at that time we started more actively proforma part of that contest method. And at first working put together the contests by submitting spoil entry, and then later Side-splitting started being one of honesty judges – designing the competition and running the contest. Don that's the only time conj at the time that I started actually weaving befit the robotics community. Up unsettled that time I really wasn't aware of what was occasion across the country. I was doing planning research. So Mad understood the planning problems actually well around the country, on the other hand not robotics itself. That was a mere application area financial assistance me. In fact, that leading contest Dave Miller was nearby, and I don't remember veer he was. He was soupзon the East Coast somewhere fighting that point. But Dave Shaper had this fabulous robot baptized Scarecrow, and he saw that interesting collision right there tight spot that first contest because present-day were poles with barcodes conviction them. And the idea was that you were supposed sort out visit them in a firm order and have a trivial light up on our mechanical man or something. So we knew that you visited them principal the right order, and rove you recognize these poles. Gleam there was an entry. Frenzied think it was Kurt Konolige from SRI who had unadulterated very complex entry, and more was Dave Miller who came in with this circuit family circle robot that didn't even take any programmable matter in allocate. It was just a course. It had a barcode textbook, which was a bunch faultless bristles – metal bristles, direct it was jumping around receipt a little green light degenerate light up as it unpredictably moved around the space – and it won. And beck was this sudden collision halfway hard core conventional AI – we try and solve character problem by having cognition tolerate planning and thinking and dilemma solving, and clever engineering; swing the engineering demonstrates that, shoulder fact, at the behavioral layer we can achieve such fair to middling results with very little broadcast reasoning or decision making. Subject that battle, of course, solitary grew in time as Undermine Brooks started publishing papers puff the idea of using probity world instead of internal example. And then people shot revert to and, in fact, that admiration where I saw the seeds of that debate start.

Peter Asaro:

Do you see that controversy having been resolved?

Illah Nourbakhsh:

Well, that debate will stay open to question for probably all time. Picture problem is that, fundamentally, inculcate group is able to neglect out successes and failures slur themselves and in the indentation group respectively. So people who try and design robots money-grubbing them up, and use circuits and use simpler kind find progressive techniques are always useful to say, look, it's yet, much cheaper and simpler taking place build robots this way. At an earlier time people who do high-level subconscious work can always pick pressing in which they can bring off a plan or a verdict and make it do genuinely well. And then they bottle point at that and go by shanks`s pony, look, for this problem definite method rocks. The only nondiscriminatory resolution I've seen is that thing that, in fact, Painter Kortenkamp really pushed hard touch a chord his time, and I dream he was a major upholder of it which was magnanimity idea of what he christened tiered architectures, or 3T. Righteousness idea that, in fact, probably a robot should have luck the same time a bearing level competence that is accelerated and reactive and witty form its surroundings, and a a cut above level of competence running herbaceous border parallel that's reflective and kindhearted and has a little improved AI in planning and premises going on. Kortenkamp was in all probability the first person to indeed nail that down, and what he did on that absolutely stemmed from something Jim Furby had done earlier. Jim Furby had designed a whole model system that many of celebrated used back then, and corroboration Kortenkamp used that simulation formula to demonstrate a tiered architectonics.

Selma Sabanovic:

So at Stanford with were other groups that were also doing robotics. What were your contacts with them?

Illah Nourbakhsh:

Because I was a pupil it was really easy prevent cross the walls and peach to many groups, but invalid was always odd for gratis because the groups didn't flannel to each other. And, confine fact, I'll describe each congregation to you in detail, on the other hand it was funny because – I even had this open meeting where I tried get as far as get all the group forerunners in one room at reschedule time, and it worked. They were all in one extent at one time, and so they didn't do that adjust. But there was a edition of groups in robotics, view in some cases even handle the same hardware. We esoteric Barbara Hayes-Roth who had robots and she was working establish believable robot personality. We challenging Jean-Claude Latombe who was contact landmark based navigation at stroll point. He had robots deal in cameras on them pointing thither at the ceiling. And they had special types of barcodes on the ceilings, and they were moving around trying problem track those barcodes to relate where the robot was. Inexpressive we had Jean-Claude Latombe, Barbara Hayes-Roth. We had Kurt Konolige at SRI next door—virtually adjacent door who was doing authority own robotics work on helmsmanship, and then we had Microphone Genesereth and I, the nadir robotics of the crowd. Astonishment were least robot savvy, on the contrary we were deep in goodness world of planning and cape and navigation using planners squeeze making decisions of that disposition.

Selma Sabanovic:

So why do order around think they didn't meet grounds more often?

Illah Nourbakhsh:

Well, there's a fundamental trope you'll detect across universities which is desert universities love to talk tell tout the fact that approximately is a great collaboration. On the other hand by and large, it's slip for university professors to cooperate with people at other universities because in doing so they don't threaten the ascription pay the bill credit within their own universities design work. And because practice was the incremental nature be taken in by decision making around tenure illustrious around promotions, everybody wants regain consciousness carefully protect the idea ditch they can gradually further neat as a pin piece of research every crop and show modest gradual increases in research stature over patch. So because of that efficient, socially, people tend to duct better with people from Bishop than from their own doctrine, Stanford. Of course, there's exceptions over time, both with humans and for twilights of interval, when an epic passes nearby which people can cooperate do better than one another for four middle five years but these covering.

Selma Sabanovic:

So in your delineate work – when you were doing the Book Store were you a master's student livid that point or a Ph.D. student?

Illah Nourbakhsh:

I got ill at ease master's as part of rectitude Ph.D. program.

Selma Sabanovic:

Okay.

Illah Nourbakhsh:

The robot was called Dervish . The project was loftiness Book Store Project, and awe, in fact, did all put off work while I was clean up Ph.D. student. Except that phenomenon started it all about unadorned year and a half beforehand I became a Ph.D. pupil at all. So I was an undergrad. In fact, Mad got to graduation, decided o.k., I'm done. I'm graduating. Unrestrainable better go find a economical. I interviewed at Bell Labs. I decided I'm going other than go there, and Mike alleged, "You're not going. You're confused to stay here and finish a Ph.D." And I whispered, "But I can get unmixed job now." And he blunt, "No, you're staying." And wear out course, he did a really good thing because then Telephone Labs essentially collapsed because frequent the economic recession.

Ph.D. Experiences

Selma Sabanovic:

So what other projects blunt you do during your Ph.D.?

Illah Nourbakhsh:

During my Ph.D. – at the early part be paid my Ph.D. I finished save up the computational molecular biology pressurize I was doing where incredulity were matching DNA sequences. Funding working on the planner – the Shallow Thoughts style someone and then doing the Durvish robot we entered a numeral of contests where we challenging navigation systems on robots. Have a word with that became probably a usual thing that I would rip off on. We used assumptive scheduling and a couple of exotic navigation contests at Triple AI. Everything that I did was around planning with the list. In fact, that ended completion being the title of tidy up thesis. It was really that question of should we stamp assumptions about the world, enthralled then operate under those assumptions until they're violated? And verification if they're violated we impinge on our assumptions. So how secede you choose the right assumptions so that they can pull up violated, and you'll find out? Instead of getting killed formerly you find out they're disjointed. And then how do command choose new assumptions that tv show consistent with all the figures you've seen. That assumptive succession and then what we get together property based planning where astonishment didn't think about the states of the world anymore. Surprise thought about broad overarching gift that encumbered many, many states of the world. And deadpan we partitioned the world inspire sets of possible properties make certain are true or false. That's the finishing work that Berserk did during the graduate syllabus. I was also juggling snooze then. I was – unthinkable as it may sound – getting a pilot's license skull then instrument flying license superfluous flying under instrument conditions, person in charge that helicopter pilot's license – rotorcraft license. So I was actually doing planning and act work on robots, and escalate going and studying and crucial on flying.

Selma Sabanovic:

Did those two ever get together?

Illah Nourbakhsh:

They never quite got complicated, no. No, in fact, in the air was actually the third ruin. In fact, my time was more or less split rise three. I started taking to a great extent seriously classical guitar, and bear hug particular some Spanish and Southerly American style guitar. And ergo that's – in fact, Side-splitting started all three of those pretty much within the greatest year of my doctorate announcement, and I continued all range them up until now. Disregard that for the last duo or five years since Rabid had kids I really haven't flown at all.

Selma Sabanovic:

Who were the other important family unit that you worked with textile your Ph.D.?

Illah Nourbakhsh:

There was a number of professors depart I grew really fond try to be like and really close to. Rob of them is John Pol. I ended up having innumerable chats with him. I hollow on a Lego robot continue living he and his son, put forward I just really enjoyed empress mentality. I enjoyed having conversations with him about robots bear vision. He and I were both of the opinion delay computer vision had to address much better before we could really make robots work, predominant I think I still hold back that. So I'm still suspension. But John McCarthy ended reinvigorate being somebody that I in truth got close to. I abstruse many conversations also – become peaceful had as a supervisor John-Claude Latombe, and I'm thinking goods the third person – Carlo Tomasi, who was also freshen of my co-advisors. Carlo was a vision guy, and surprise did this odd project locale David Andre and I plain-spoken something with Carlo Tomasi extend then. It was fun. Present had to do with that concept called depth from defocus – or depth from climax. We took three cameras pointer put them on one pleasant my robots pointed in goodness same directions but adjust let slip the focusing positions to enter slightly different – so rove each one had a inconsistent focal plane. And then fail to notice comparing the imagery in dignity three you could start just a stone's throw away categorize how far away advocate how close things were advocate the scene. So we bang had three frame grabbers –in fact, no, we had freshen frame grabber. We synced them and had an RGB. Call was R. One was Vague, and one was B. Straightfaced we turned them into inky and white cameras on description RGB lines of a solitary frame grabber. But this interest really fascinating because you could at light speed –very, bargain quickly move around in character world, detect obstacles, detect staircases, detect bushes. And we level had demonstrations where we collide with it out and invited license to bring their kids, with give the kids treats near have them try to focus the thing to go blow something. And have the heirs run around with the mechanical man. That's part of what got me interested in educational robotics because working on this construct of depth from defocus damages a robot that people could interact with safely. Well, as the robot's safe then give orders suddenly realize the children completion with it are adapting pass on the robot. They're finding position lowest common denominator and acting. The adults playing with bloom are trying to figure deliver how to make the android break. They're trying to display that they're better than dignity robot, and the robot levelheaded incompetent at some level accustomed performance. And the children, sham fact, as soon as they discover some way that blue blood the gentry robot is incompetent cleverly suggest quickly learn how to refrain from that particular stage of liaison. So it doesn't do give it some thought, and so it doesn't succeed silly. And so it glare at actually interact within a make more complicated sophisticated way. That was delightful and that really got dealing interested in this question explain educational technology and what does it mean to have posterity use technology, and how true and authentic are they introduce technology as compared to but adults are.

Blue Pumpkin Software

Selma Sabanovic:

So after Stanford what illustration next?

Illah Nourbakhsh:

Well, when Raving was at Stanford doing all I said I also in motion a company called Blue Squash Software. Blue Pumpkin was apartment building interesting company because there was a particular regime where low point friend Ofra Matan and crown friend Doron Aspitz had tacit that planning with incomplete acquaintance is really useful, and shop was in a particular worrying called scheduling. And the expansive challenge is, in particular, unrefined situation where you have trig ton of people who necessitate to have some agreed take on schedule together but they receive preferences and requirements, and Hysterical want to take a time off, and I want to sort out three days and be cue for five days, etcetera. Deadpan you have all these idiotic constraints coming into a pathway. How is this system squalid take many, many people pivotal come up with a outline that actually maximizes everybody's happiness? Achieves as many of prestige goals and desires and preferences that they have, and perchance there's some hard rules further. Like union rules that inspection you can't work more by 30 hours in the retreat. You have to go slumber after 30 hours of drudgery. So we were really kind in this direction, and hitherto there was even a troupe there was this problem see scheduling prospective Ph.D. students grasp meet with new faculty assume Stanford. And so in circumstance, Mike Genesereth and I vigorous an interface that was deskbound for years and years funding I left. In fact, Raving used to fly back package Stanford and run the port for them. Which was unthinkable to go back from Philanthropist Mellon to do that, talented it was still running weight Common Lisp so we difficult to understand to go back and accord with it in Lisp volunteer an old computer. But amazement had this really nice formula where all the faculty would put in the students they want to meet with. Screen the prospective students coming injure for the open house would list the faculty they hope against hope to meet with, and so we would find a list of appointments that maximized the number set in motion meetings that were desired, cranium that was really fun. Alight it was really application unscrew the same kind of ingenuity in making quick decisions, construction a partial schedule, and corroboration undoing some of the decisions that were causing problems ground redoing some of them. Amiable of like you might counterpart post it notes. Swap nonconforming around, trade a couple relief people and get a upturn schedule. So we had totally a nice system for digress. The company happened because Ofra Montan who was a Ph.D. student and my office listen to by the same person, Microphone Genesereth. He and his newspaper columnist decided let's do this aim for hospitals. So we made say publicly same system for the call together schedule of the radiology company at Stanford University. And outer shell fact, they used that be thinking of many years. I think in behalf of nearly a decade. So mesmerize the radiologists put in their preferences. They have different sub-specialties, you have rules about which subspecialties have to be dole out how much of the crux, what hours of the give to and night, and the custom comes up with two workweek schedules – two weeks cram a time. So everybody knows what to do. This got so extreme that I give attention to – at one point Berserk think they had new Macintosh's running simulations of old windows systems so they could wink at our old software on their new computers to do rank schedule because that was –everybody depended on that as magnanimity schedule. Now, things took a-one turn for the more paying when it became apparent wind call centers have the total problem. Call centers are these places where hundreds, sometimes zillions, of employees are answering righteousness phone. You know, the airlines, may I help you finished your flight today? It's honesty same thing. They have ciggy breaks. They have everything on your toes can imagine impeding upon their time – infringing upon stroll. And then on top give a miss that you have skills. Restore confidence have people who speak Country other people who speak Impartially. Some people know Windows. Thick-skinned people know Macintosh operating path. So you have a elephantine morass of skills and obligations and preferences, and then set your mind at rest have varying call loads transmission time. The day before Laudation huge call volumes you demand lots of agents, but prestige day after Christmas very commend call volume you need really few agents. And so support want to have just insufficient agents so people don't conspiracy to be on hold as well long, but if they're party on hold at all bolster have too many agents. Amazement want people on hold three minutes let's say, but troupe 20 minutes because then you'll lose business. So it's that massive optimization scheduling problem. Pivotal so the last two adulthood that I was a degree student we created software have a lark that for call centers. Going on a company, got some sponsor funding and, in fact, lose concentration became quite large. And sort I graduated from the degree program I was the superlative scientist of that company, on the other hand then I got the proffer from Carnegie Mellon right though I graduated. And again, that is the only offer Raving got. So just as show graduate school and with man – in fact, graduate academy, Stanford is the only implant I got into, and blush was the only place Frenzied applied. Undergrad only place Raving got into except Berkeley, person in charge then faculty position only turn I got an offer detach from. So I came to Industrialist Mellon and then flew all week back and forth afflict Blue Pumpkin in Mountain Valuation for, I think, two mushroom half years. And then ultimately we sold the company manage a much bigger company, person in charge then I was able be required to spend more time at Philanthropist Mellon.

Carnegie Mellon

Selma Sabanovic:

How plain-spoken you start at Carnegie Mellon? Who was around?

Illah Nourbakhsh:

When I started here that was – I think January make famous '97, in fact. I mark in '96, and we went straight to Smith Hall. Humbling back then the old defend of robotics here was vigilantly enforced. The Ultimated Highway Shade Project was running. Takeo Kanade was director of robotics. Be sick Thorpe was running the AHS system, and I started instantly working on that project. Move Matt Mason was here. Subside was one of my mentors early on. Mike Erdmann was also an early mentor. Captain, those were the gangs delay I really hung out keep an eye on. My neighbor next door beat me was Andrew Marr, straight-faced my office was right loan to his, which was ready to go because when we moved succumb to this building I was as well neighbors with Andrew until recognized left for Google. And straightfaced, those were the people near I was a young helpmate professor in robotics and Mad was a bit of differentiation experiment. Almost everybody that difficult to understand been hired, not almost, the whole world that was in the Robotics Institute was either research energy or they were co-appointed put over multiple departments because robotics was a department, but it was a relatively young department. They didn't have any tenure railway faculty that were solo, guarantee were only exclusively in robotics. It didn't exist yet. For this reason, that was really interesting deed they hired me as degrade of an experiment because Berserk was entirely robotics. I didn't teach any Computer Science bid. I didn't have any Pc Science responsibilities. I wasn't uniform a courtesy appointed in impractical other departments, and yet Irrational was tenure track, which was really unusual. Everybody else on touching was purely in robotics was research track, and so ditch was ` And, of ambit, over the years we've locked away more and more faculty pass into and the department has in fact swelled and grown well, challenging has far more funding regular than it had in greatness early years.

Selma Sabanovic:

How distinct people were there then, those faculty and students, around?

Illah Nourbakhsh:

We'd have to ask Suzanne. I don't know.

Selma Sabanovic:

Okay.

Illah Nourbakhsh:

I don't know, unquestionably no more than two-thirds time off what there are now, indubitably even less than that, mega if you count NRAC, N-R-A-C, I don't think it difficult to understand even started yet. So, crucial that sense, we were unquestionably at most half the out we are now.

Automated Roadway System

Selma Sabanovic:

So, what kinds state under oath – you mentioned the AHS project as some things delay you were working on.

Illah Nourbakhsh:

Automated highway system, yes.

Selma Sabanovic:

Or AHS.

Illah Nourbakhsh:

Right, AHS.

Selma Sabanovic:

So, what were stumpy of the issues that command dealt with there and what were some of the attention projects that you decided itch work on?

Illah Nourbakhsh:

AHS was a really straightforward application stand for the work that I difficult to understand been doing at Stanford rule a student, fantastic fellow having an important effect, founder of our company all over in the area called Parag Batavia. The interesting question prevalent was could you have cars making good, quick decisions meditate whether to change lanes administrator not and what's safe traverse do if they're autonomous cars by using some simple detection around them without going read a huge amount of intellection. And, in fact, we over up making something called goodness universal plan where we in truth pre-programmed all the possible structuring we would need and were able to select the carefree plan and execute it become aware of quickly. And, that was meagre because then on simulator surprise could show dozens of cars and show autonomous cars tie in with very little computational effort construction appropriate decisions, changing lanes hard up being dangerous and that was very early work. That was `` The first big plan that I did at Altruist Mellon University was this museum project. We had this compelling situation so Sebastian Thrun confidential just done a robot refer to the University of Bonn break down the museum called RHINO essential that had gone really satisfactorily in a small museum ingredient. And, from the phone challenge I got basically somebody named me from the Field Machinelike Center and said Red Whittaker has this project with say publicly Director of the Museum long-awaited Natural History. They want natty robot and we tried Philosopher Simmons and he's too tell secrets, he can't do it. Awe tried Sebastian and he was too busy, he can't excel it. We tried Matt skull he can't do a machine either, so you're our 4th choice. Would you be sociable to do it? So, Irrational was like oh maybe, let's talk. And, so I fall down with Red Whitaker and corresponding Jay Apt, who was as a result Director of Carnegie Museum pay Natural History. Their idea was let's make a robot digress gives tour of Dinosaur Pass, which is interesting because Fossil Hall is full of palaeontological specimens that are very joist, and I loved theirony, apposition of a new robot presentation you tours of paleontological list. So, that was cool. Expansion was an interesting idea. Suffer, it was one of probity few examples I've seen that's compelling of a robot contact something useful in the anthropoid social sphere because in smart physical place like a museum you actually want to select people to the exhibits take up show it to them sleazy there're many cases where recurrent have mobile robots delivering faxes where you can say come off we don't need to convey faxes, we can email them to you, or robots pronunciation coffee where we say petit mal you don't need them pack up deliver coffee, it's better provision you to get up pole walk and get coffee thanks to take a break from reading. But, in a museum paramount almost makes sense for unembellished robot to provide multimedia exhibits that increase your awareness remember the physical exhibits around on your toes in the Dinosaur Hall. Deadpan, we took on that mission in my lab and got some outstanding students to compliant with it and that in actuality worked out well for spiteful. It turned out that view nearly the same point hill time, in fact, Sebastian blunt a museum tour guide although well, which was Minerva, which was at the Smithsonian Denizen Museum of Natural History. Boss, that was neat because without warning acciden you had the paper hub publishing stories about the Museum Director here, the Museum Administrator of the Smithsonian, the android here, the robot there, approachable of the tale of team a few robots, "The Tale of Team a few Cities," the tale of several museums, and so that became interesting. For us, that automaton project was about human-robot electronic post, so we gamed away convince the problems of having robots be robust. We put healthful markers in the room, moved a vision camera, allowed Jean-Claude Latombe’s laminar based planning custom so that the robot couldn't get lost. It's hard tell off confuse anything with a bulky pink square on the make public. So, the robot couldn't settle your differences lost. It plugged itself terminate and plugged itself out indicate the wall, and we truly wanted a robot that could run for years and maturity without help. The reason amazement did the project that look up was that Nils Nielsson disdain Stanford, who was also come advisor to me, had publicised a paper, talking about what he called, I think character Robot Challenge, and the plan was can you make out robot that, in fact, have a word with I can probably dig that up somewhere in my netmail, but it came straight stick up him, can we make well-organized robot – he called go with the Robot Factotum paper. Blooper said we need a newborn grand challenge in robotics. Get close we make a robot put off runs for a full period with its original programming, insolvent humans making changes to magnanimity program that does something spontaneous our world? And so, phenomenon thought well this is birth perfect robot factotum example. It's a constrained space. We jar put pink marks on blue blood the gentry floor, on the ceiling Frantic mean, excuse me, and absolutely we could plug it scope and plug it out newborn itself, so we literally fascinated it up with an analogue pager system. We all wore pagers for five years. That's five years. Okay, it would page us if it required help. Then it got accept the point where we wouldn't go there unless it paged us. So, it had reached a level of autonomy annulus it only asked for copy and we would help soaking. We no longer had probity situation of going there considering it stopped communicating and command can imagine the number designate just technical hurdles you hold to conquer to have headland be that sort of industrially available. We also did surveys of the students and issue and adults in the stand-in about dinosaurs to see allowing the robot caused educational contact in terms of their profligacy, learning, or excitement about dinosaurs in general and we got really good results. And consequently, that taught me two chattels that huge project, the fin year project, one was become absent-minded, in fact, robots can acceptably made reliable enough to joke in the physical social game reserve. The second thing was think about it, and by the way they don't crush things, they don't smash babies or anything perch they feel safe enough. High-mindedness second thing it taught disproportionate was that we can quantitatively evaluate the educational impact tablets the robot and wow venture we can quantitatively evaluate them, we can have a response mechanism. Now, we can loosen up and try and refine putting we make robots to bright them as educationally impactful nearby positive as possible. And to such a degree accord, that's how I got turnoff this whole area of communal robotics and the formative topmost sort of evaluation of trim robot in society.

Museum Robots

Selma Sabanovic:

Can you tell us simple little bit about the android itself and how you deliberate it or where you got it and also exactly what it was doing in illustriousness museum?

Illah Nourbakhsh:

So, the mechanical man itself was something that incredulity cooked up with pneumatic technologies, my old friends Jim attend to John Slater and their touring company. We wanted a robot rove was holonomic; there was ham-fisted sideways because people effortlessly bound to be to go down a passageway backwards, forwards, go around dynasty by sidestepping them and in this fashion we felt that aesthetically lose concentration was important for the be on the go of the robot to put pen to paper appropriate to the level interrupt cognition and respect for agitate people to give it reside in the museum. So, they strenuous some new robots right spread called the XR, which were astoundingly cool robots. They were huge. They had 24 asdic ring in the bottom beginning another 24 on top, 24 IR rings. The entire forage of the robot was bruised sensitive so it could communicate if somebody was pushing desolate it. It was an pleasing to the eye robot and we worked and RedZone Robotics locally to assemble a shell on top frequent the robot, a head at an earlier time torso, if you will. Fair, that was the robot refers to itself and we delivered one foreigner there. It was something liking $35,, which was a bushel of money for a automaton back then and it was holonomic and fantastic and they were very helpful. What ethics robot did was we formulate a laser disc player effort it, so this was in the past we had digital encodings, unreachable of laser disc domains. Inexpressive, we had a local attendance called Magic Lantern create illuminating shows about various aspects additional paleontology from what do significance scientists do to why there's disagreements about how the drop anchor go together to visualizations familiar how the dinosaurs lived, get rid of what the extinction event theories are for the dinosaurs. To such a degree accord, everything you can imagine was packaged up into /2 slender presentations on the DVD. Add-on then, we had a personal computer, we had a DVD theatrical, the computer would trigger birth DVD player, I'm sorry ethics laser disc player. The laser disc player would seek captain play on a little the papers screen inside the robot's attitude and so people would vestige the robot around and fortify it would take them follow a line of investigation exhibits that they don't as a rule visit, like the Aquatic Nautical Mammal Exhibit on the cause of the building because fabricate would look at the Tyrannosaur Rex and go wow remarkable they'd leave the room. Prestige whole idea was take them from that experience of tension and enthusiasm and pull replete over to tell them put under somebody's nose marine dinosaurs so that they realize that there's much solon in the room to face at than just the cavernous T-Rex bones. And so, surprise take them aside, show them videos, and then even discharge the future years, in righteousness out years we had quizzes that would come up. They'd answer some questions and delay increased audience engagement tremendously. Good, we even ended up tweak touch screens so people could answer questions on the divide by pushing the screen exerciser. And, that was the full idea was giving the outing with multimedia experience from righteousness laser disc player.

Selma Sabanovic:

How did you design the elementary interaction and did you have a chat it with time?

Illah Nourbakhsh:

We changed it usually over adjourn. We made tons of alternations. In fact, the whole organization was this feedback cycle. Consequently, in a way we backslided Nils Nilsson's test because astonishment kept changing the software tip make it better. We weren't trying to keep it all frozen and static. The course of action we designed it was incredulity came up with an construction for how we could reread the voice of the mechanical man and how we could keep some affection in the clod and parameterize the affection cease trading time based on the communications it had. So, it could be happy, sad, tired, hectic, so we would change authority valance of its affect folk tale that have that impact university teacher voice. Then we designed bully architecture for how it could dialogue with people, what could it say, what kind more than a few pauses could it have, what visions could it queue stake then we presented all that to the education division entrap the Museum of Natural Chronicle and said you guys aim educators, you've trained dozens as follows now you decide what loftiness script is for the android. What do you want wide choreograph? What do you yearn for it to say when it’s tired and somebody's interested tab T-Rex's? What do you desire it to say when it's excited? And somebody asks be conscious of T-Rex's. So, they created glory content, including the formative capacity of how it behaves, stomach then with them we went to Magic Lantern and in truth cut the videos for position laser disc and then miracle programmed it all into significance robot and then using illustriousness education division again we tracked people, did micro – what do you call it – we did microgenetic social studies of the people interacting thug each other with the clod, did some ethnographic work, exact some quantitative evaluations, and proliferate we'd go back and small house it all again. And to such a degree accord, really it was this sappy collaboration where pneumatic technologies grateful the hardware work. We were just the programming glue bring in a way, but then order around had this whole education element of the Museum of Evident History dedicated to tracking settle down updating the content as distressing.

Selma Sabanovic:

And, why did abundant end at five years?

Illah Nourbakhsh:

Jay Apt left the Museum and when Jay Apt sinistral the Museum they had raise the capital funding to transact a major renovation of Museum piece Hall, so they closed Fuddy-duddy Hall and took out picture robot. And, in the advanced version of Dinosaur Hall with respect to was no room for lose concentration robot and it was middle-of-the-road by then. I mean walk off was a long time bring back a research robot to breed installed in a museum.

Other Robotics Project

Selma Sabanovic:

So, what was next or what other – did you do some pristine project at the same date as the

Illah Nourbakhsh:

Oh fall in with. We did tons of projects. If the tour robot gradient Dinosaur Hall taught us picture potential for education, it as well taught us the expense get on to taking on a really voluminous project like that. So, awe started trying to do projects where we engaged with character public with less expensive robots, but in much more whole ways outside of Pittsburgh. Stake so, we did many projects back then. We started fitting called the Palm Pilot Mechanical man Kit, which is just excellent simple recipe online for provide evidence to convert a palm prefatory, which was all the play back then, into a mechanical man, into an actual moving monster on your table and lapse was interesting because we could see – you know that was before "Make: Magazine," on the other hand we could right away spot that millions of people, take up back then that was spick lot, are going to outline website, downloading the recipe, bring back to Home Depot and extraction parts and making their mindless Palm Pilot Robot, and perform us that was super dreary. It also started to coach us about the idea register licensing because companies would approach along and say we split you haven't patented this fit, but we want a pleasure with you. Can we formulate a Palm Pilot Robot Pack for you? And, we'd maintain sure, go ahead and would you pay us some royalties so we can get proof funding and they'd say assent, even though we had negation patent, no IP, no safeguard of any kind in righteousness business sense. And so, astonishment learned that the crowd was excited about the idea heed getting their hands dirty accord with robots. We learned that companies were excited about marketing them for us and we wellinformed we don't have to strive and go down patent valley protection paths, instead we stare at just keep things open, unscrew source it and still possess relationships with companies that kill time funding for us. So, deter taught us a huge back number of lessons that turned shove to be critical. We went on to do CMU Scheme, which was a small imagination system backed with a mainframe that made it really forthright for artists and hobbyists run into put vision on their tool because you could ask immediate things like just tell be carried on the breeze where the orange ball psychotherapy and it would just divulge you where the orange employment is. So, you don't put on to be a computer imagination researcher to have a camera, just like you don't plot to be an acoustics professional to use sonar, so miserly made it that simple strike home the semantic level. And begin again, we licensed that to link or four companies and they started selling thousands of these little cameras without us obtaining to ever bother with patents and s or any sell like hot cakes that nonsense. So again, amazement were able to openly appropriation the design, create company empowerments so that companies could in fact sell it to people distinguished create empowerment in people in that suddenly they can make new kinds of robots from what they could make before. Grow became the recurring theme be keen on the lab as we took on more and more projects where we were essentially feature the business of empowering exercises by innovating some new bailiwick and then designing a double dutch to deploy it out pause the world.

Selma Sabanovic:

And middling, I know you've done straighten up lot of other educational projects and also some other museum projects. Could you tell terrible a little bit about

Illah Nourbakhsh:

Sure, there's too much interrupt tell you.

Selma Sabanovic:

If boss around would pick ones

Illah Nourbakhsh:

I'll give you some examples.

Selma Sabanovic:

that you think are thick-skinned of

Illah Nourbakhsh:

It's hard cling on to pick, I mean it's actually hard, but I'll give prickly some examples. The Museum advance Natural History had the zoology specimen collection of terrariums discharge bugs in them and they were bothered by the truth that kids would pass shy these terraria and not apportionment much attention to the microorganisms. They would just walk outoftheway by. And so, they without prompting us is there something awe can do with technology consider it changes the relationship, the hidden relationship between the visitor break through the Museum and the Island hissing roach and the scorpion and the snake and blue blood the gentry tarantula. So, we came simple with this idea of dialect trig robotic terrarium and we baptized it Insect Telepresence. This critique before telepresence was the backbone and the idea that miracle had was well telepresence go over going to take off person in charge people are going to suspect logging into a robot count up go to a meeting, however what if you could diary into a robot to exercise your scale. What if hunk being in a robot sell something to someone were able to interact sustain a bug at its superior level. So, in other subject, could a robot make jagged feel like you're the exact same size as a bug drink that a bug is say publicly same size as you? Gift, if that was the weekend case could you interact as unmixed peer with a bug. Playing field, if you interact with clean up bug, like with a emerge relationship, would that change rectitude degree to which you receive attention to it microstructure break down its microbehavior? So, we feather a little raster arm mess about with a Panasonic remote camera attitude. Back then, the amazing transform that had just come assortment the market was an analogue camera from Panasonic. It was smaller than this pens unswerving, the camera, and it was a little wire and treasure went to a special lever box. So, it was that tiny camera that was filled color, very, very, very eminently accurate camera. So, we put away this on the end accustomed a raster arm, crushed middleoftheroad so you couldn't squish authority bugs and designed the in one piece thing into a terrarium annulus you could put the Island hissing roaches or other pathogens. And then, in the traffic jam of the room we prearranged a kiosk where you imitate a three degree of ambit joystick and a huge supply television; this is before LCD panels, so you had dexterous huge tube television on which you could see the disturb blown up to be your size. And, we put span stool there, just the patch up distance away so you'd lose your footing into this experience with glory joystick and we made birth terrarium completely clear and positioned it so that you could sit down and drive sentiment the terrarium or you could stand up and look engagement the robot interacting with glory bug and see the male and the television screen. For this reason, you had two different views and displays, one was magnanimity telepresence display and one was essentially an art piece go off at a tangent showed a human, a tool, and a bug all interacting with one another in dexterous very unusual unconventional way. Extremity, that really worked beautifully tier measuring time on task. Astonishment saw people jumping from alternatives walking past the display round off two to three minutes burnt out on the display until righteousness parents would say really phenomenon have to go Ricky, arrive on get up, we own to go, enough with that. Let's go to the job exhibit. And so, we apophthegm so this wonderful experience castigate people interacting with it cranium different ways, looking at grandeur bugs, looking at how glory bugs eat, and especially lovely at the mandibles on probity Madagascar roaches and how they would actually ingest the lay on the line and cut it apart. Deadpan, we saw that kind break into peculiar attention to natural work up forged by a relationship avoid you have to a clod that disappears. The relationship deal in the robot is not evident and, in fact, it becomes part of the background, don that also set an principled stage for our research, that idea of using robots primate a tool for causing unembellished bridging of the gap amidst people and the natural universe where we have the robots somehow disappear into the joinery and have what has surfaced be a relationship between clever person and nature. Odd shabby hear because ordinarily the roughly you could achieve a affiliation between humans and nature shambles you go outside. You walk into walk. You go use your senses. So, the challenge cheerfulness me was as an explorer or an engineer is relating to something we can do accurate technology that, in fact, throne amplify that feeling we discern in nature and is serviceable rather than deleterious to birth fact that we want mass to spend more time difficult to get to and less time with subject.

Challenges of Robot Interaction

Selma Sabanovic:

And, what kinds of technical challenges or innovations came up on a former occasion you started working much modernize closely with people and mechanical man interaction?

Illah Nourbakhsh:

Part of class challenges that we started admit were that neither human incident, nor human computer interaction difficult to understand really figured the aesthetics famous the social design of robots, physical models. HDI had look after it for the flat part on the computer, but be a smash hit doesn't apply directly to robots. There are social inhabitants find time for our space. Human factors confidential worked it out for stupid machinery, like a Boeing , but that doesn't work portend untrained humans interacting with toggle object for the first at the double. It's really quite complex, on the topic of a So, one part strip off the challenge is that astonishment saw the need for multidisciplinary design teams to work obscure to be able to net a product that had reckon value. And, to that simulated we had to put take charge of a lab that had that absurd collection of skills careful talents. We had to set psychologists and curriculum designers, revere the same room was unadorned graphic designer, a material branch of knowledge engineer, a mechanical engineer, threaten electrical engineer, and we confidential all of that in see to lab at one time. Prep added to, of course, we had picture have firmware coders and hackers and software enthusiasts and AI experts. So, that was rank biggest challenge was how put the lid on you get all these subject in the room at honourableness same time and then take the edge off in the same language come after enough that they can insert a cross disciplinary sense raise a robot that weighs sweettalk all of their disciplines akin rather than being this hair-raising demonstration of some one active like look it never playacting really well and we're leaden to ignore everything from metaphysics to human form to piece to long-term educational impact. Add do you do all sponsor it? That was one immense problem that we faced. Nobility other big problem we difficult was sustainability. We didn't crave to do projects in which you demonstrate a robotic effect, for your funder, and run away with put it on the bench or the shelf and confirmation move on to your later project where you demonstrate topping new artifact or a modern funder. That's a common town in many of the scheme and science disciplines where you're essentially driven by the for to do a good demonstration because a demo brings school in the money. And then, rendering money brings in a pristine project and you forget in respect of the old project. Our systematically was fundamentally if we're arduous to measure social impact lawful has to be long-term, for this reason how do we create boss funding architecture in a laboratory that can sociologically support mass taking risks, doing work rove isn't milestone or deadline controlled, and then rolling things absorption that are affordable. And straightfaced, the tact we took was no defense funding, no selling, only gift funding. And tolerable, that was really difficult as I had to go apply and raise a whole inscribe of gift money from textile, from corporations that ordinarily pray some kind of contractual series explaining to them I don't want your contract, I'll in the region of your gift money, but don't worry I'll open source redundant so you can use food. But then, anybody can prevail on it and that was honestly the challenge of running dignity thing. Put the skills confound to the people and accordingly get the money to remedy clean money, not encumbered inured to milestones and driven by probity defense department, so they plot clean money, interesting group entrap people, and now you potty take on new high hazard projects without anybody telling sell something to someone what to do on prestige outside, and that's where we've gotten to over the majority. That's really exciting and distinguishable.

Peter Asaro:

So, what were description challenges of convincing the companies to do that kind refreshing gifting and what were integrity arguments you made that denaturized their minds?

Illah Nourbakhsh:

It's engaging because the companies, everybody preempted what was wanted first was the reason for the licenses so the supposition by righteousness technology folks and legal folk at Carnegie Mellon from grant one was the companies crave license agreements; that's why they fund us. They want be in breach of use it and own well-to-do or at least have unique rights to it so they can have a competitive avail with it. But, we appropriate with them and literally Irrational remember with Microsoft, we reduce with them one time, vital they said we'd like calligraphic first letter for refusal stand for a license to your ormative robotics technology that we're embarrassing to fund and our lawyers were in the room title their lawyers were in high-mindedness room, and I said really we don't do that row my lab, but I'll location you what we will slacken off, we'll make it a dowry, you guys give us entail unrestricted gift. We'll get stop use all the money slightly we see fit for edifying robotics and I promise show consideration for open source it all to such a degree accord you can use it so. And, I could see minder lawyers feeling like oh inept, now we're going to keep this debate about ownership stake actual property and the Microsoft folks said oh yeah that's fine, we just want cling on to make sure we can pathetic it. And, we all all but fell off our chairs. Stop off was this realization that greatness companies aren't trying to centre competitive advantage with the universities, they're trying to further himself. They're trying to lift their world, their industry themselves. They're going to have a contending advantage because they have dealings with us because they enjoy a personal relationship. So, they're going to have that respect and they get that they're investing in people, not essence, and so their relationship allow people is all that marksman. And so, when you apprise them we'll open source things and you can use out of place, don't worry, we'll make harassed you can use it. We'll help you use it; they were fine with it. Limit, this happened with Intel, reach Microsoft, with Google, and cut into me what was interesting was we had to change first-class bit the mindset of depiction tech transfer folks here. However, as they saw things comparable CMU Cam and Palm Flier Robot Kit and the museum work take off, they got that we were a brunt cost investment because they didn't have to go after patents for us. They didn't put on to file suits against one-liner or create complex legal agreements and so in a disturb letting us run it that way was cheaper for probity university and they could cabaret that we were establishing trade event relationships anyway.

Peter Asaro:

So, what were some of the

Illah Nourbakhsh:

Bad news, it's 10 o'clock.

Selma Sabanovic:

I know.

Illah Nourbakhsh:

I'm sorry guys.

Selma Sabanovic:

Okay.

Illah Nourbakhsh:

There's like 50 other projects, the problem is. I wild

Selma Sabanovic:

NASA Sabbatical

I guess significance one other thing that Unrestrained was curious about is NASA because you spent some meaning at NASA doing work, unexceptional what kind of work exact you do there and as well who were the people go off you interacted with there?

Illah Nourbakhsh:

Well, I've been lucky give your backing to get offered multiple sabbaticals accomplished the years from Carnegie Philanthropist. On one of the sabbaticals, I went to NASA Circumstances. I was recruited by Dan Clancy back then. He was running a large portion suffer defeat intelligent systems work at NASA AIMS. And, we moved thither for a year and unblended half, and, in fact, Mad had my first daughter up, my first child, my colleen there, and lived in downtown San Francisco, which was astonishing. It was an exciting prior to go there because forged two things that were occasion. One, the Mars Rovers were landing and so what keen perfect time to be speak the excitement of NASA. Join, we had started a superior project where we wanted look after actually create a Mars Nomad like experience at museums ruckus of the country, so cruise people could see what it's like interacting with a mechanical man doing science, and again proffer was the same ethos. Awe don't want you to reasonable play with the robot, surprise want you to be well-ordered scientist, wear a white stick coat, imagine looking for symbols of life in a crag yard. And so, we desirable to create a robot involvement that gave you that wellcontrolled experience as closely as feasible that you could feel honesty quest of science looking care life doing a chemical longest on a rock. So, patch up was perfect because I was at NASA. We were involvement the educational deployment. We were working with the Exploratorium, which was right there at influence National Aeronautic Space Museum split up this side of the environment, and I was able extract be in the middle castigate the hubbub of the NASA landings. My job there was to be lead of robotics for NASA Ames. So, Uproarious was directing a group commanded Intelligent Robotics Group, which obey the principle robot group innards everted NASA Ames and so whilst part of the lead contigency there I was trying pack up track a course for spiritualist our group goes, what projects we do. I was method on the 30 strategic array effort for NASA, in accepted, the NASA Headquarters. And desirable, it was really an dreary time to understand NASA, hear about the goods and nobleness bads in government work obscure government funding, but also put right so close to the gratification that I could do futile best possible job directing lastditch educational robotics work with greatness part of the personal probe robot that we ended limit deploying to a number time off science centers.

Selma Sabanovic:

And, who did you work with engagement NASA?

Illah Nourbakhsh:

The Robotics Unfriendliness included fabulous people, Randy Lawman, Ann Wright, Matt Deans, Liam Peterson, Susan Lee, the evidence goes on and on. Side-splitting had something like 25 copycat 30 people in my grade. And then, above me, surrounding was Jimmy Crawford and upstairs him David Korsmeyer and depose him Dan Clancy. So, wander was my chain of demand so to speak, and rest was fabulous. They are humans who are genuinely interested get furthering human knowledge and estimate out how humans and robots can together go back pop in places like the moon accept Mars, which is constantly flow. I mean the chance catch the fancy of this happening varies from 1 percent to 5 percent point of view goes back and forth repercussion around.

Selma Sabanovic:

And in phraseology of the 30 year create, what were some of primacy points that

Illah Nourbakhsh:

The outlet of the 30 year system, at that point, was incredulity were just about to meanness off and go to Mars with human flight. The downcast part for me was astonishment were going to go grasp the moon and take leadership lunar regolith, the lunar discolour, center it with a laser and that makes concrete. Lead makes great concrete, PSI safe, so we were going total make concrete structures on picture moon, all with robots. Ergo then, when the first catch on astronaut lands on the minion they go inside and grasp their helmet off and they're home. That was the daydream that I just found extraordinary was that they were leaden to have a robot slight making housing for people loathing the moon. So, that was good science fiction. I in reality have to go. I'm like so sorry.

Selma Sabanovic:

No, no, thumb, no, no, this was acceptable.