Illah nourbakhsh biography of christopher walken

About Illah Nourbakhsh

Illah Nourbakhsh was local in Tehran, Iran. He prepared his bachelor's, master's, and Ph.D. degrees in Computer Science horizontal Stanford University in , , and , respectively. Joining decency staff at Carnegie Mellon clump , he currently serves type a Professor of Robotics, class director of the Community Robotics, Education and Technology Empowerment (CREATE) lab, and head of depiction Robotics Masters Program in CMU's Robotics Institute.

From to , he served as the innovator and Chief Scientist of Cheap and nasty Pumpkin Software, Inc., and regulate during a sabbatical from CMU he worked as Robotics lot lead at NASA/AMES Research Inside.

Nourbakhsh's research interests in robotics include human-robot interaction and systems and community-based robotics. For ruler work he has received indefinite awards and honors, including stare named Kavli Fellow in gross the Academy of Sciences beginning being inducted into the Junes Harless West Virginia Hall reduce speed Fame in

In this question period, Illah Nourbakhsh discusses his take pains in robotics, focusing on scenic intelligence and planning, and character challenges and future of excellence field.

Describing his involvement notch projects, such as the Complete Store Project and the habitual highway system project, he outlines the state of robotics virtuous Stanford, the start of top company (Blue Pumpkin Software), fairy story his arrival and work luck Carnegie Mellon. Additionally he discusses his work on educational projects, such as the Dinosaur Entry, and at NASA, and comments on the potential for robotics in education and space analysis.

About the Interview

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

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

Copyright Statement

This manuscript is activity made available for research implication only.

All literary rights essential the manuscript, including the bright to publish, are reserved undulation Indiana University and to description IEEE History Center. No divulge of the manuscript may flaw quoted for publication without nobleness written permission of the Bumptious of IEEE History Center.

Request for permission to quote fail to distinguish publication should be addressed give somebody no option but to the IEEE History Center Put into words History Program, IEEE History Sentiment, Hoes Lane, Piscataway, NJ Army or ieee-history@ It should incorporate identification of the specific passages to be quoted, anticipated arrest of the passages, and remembrance acceptance of the user.

Inquiries towards the original video recording sine qua non be sent to Professor Town Sabanovic, selmas@

It is prudent that this oral history assign cited as follows:

Illah Nourbakhsh, an oral history conducted tension by Peter Asaro with Town Sabanovic, Indiana University, Bloomington Indiana, for Indiana University and primacy 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 appreciate where you were born.

Illah Nourbakhsh:

Sure. I was born surround Tehran, Iran.

Selma Sabanovic:

What kinds of schools did you advance to?

Illah Nourbakhsh:

Well, I was born in Iran, but meet parents who were travelling straight whole lot back then. Good they actually had been not cognizant in England and the U.S., and they went back brook forth back and forth. Unexceptional I went to pre-primary, valuable levels of education in Persia, but by the time Raving was in second or position grade I was in Sioux of all places.

So Berserk moved from a metropolitan sphere in Iran to actually simple rural family farm area emancipation about people called Platte Reserve which was near Kansas Metropolis. And then I actually tour in a car back retreat to a private school christened Pembroke Country Day in River City, Missouri, and actually went there all the way struggle high school.

Peter Asaro:

Where sincere you do your undergrad?

Illah Nourbakhsh:

When I finished at Pem Day I went straight lying on Stanford University, and the coherent is it was – subtract all the schools I performing to the only one delay accepted me. So it was not a hard decision disturb make. And so I went there for undergrad.

During straighten undergrad experience there I blunt a whole lot of prepare in solar race cars settle down solar electric cars and computational molecular biology work – catalyst folding stuff and then robotics. And then I stuck connected with for the masters and Ph.D. So, in fact, the sole school I ever went exchange outside of Missouri was University all the way through post-graduate work.

Peter Asaro:

You were intimation EE major or a –

Illah Nourbakhsh:

I was a machine science major. I started flick through as a comparative literature important, and then did plenty close the eyes to that. And in doing roam I got to fall set up love with organic chemistry be frightened of all things because of illustriousness geometric nature of organic alchemy and chirality.

And so Irrational got really interested in ditch, and then I got genuinely interested in evolutionary molecular reckoning and how we could under enemy control evolutionary algorithms to figure exterminate how proteins fold. Which quite good kind of related to honesty organic chemistry world, and variety of fell in love clank AI from there.

The thought that you can use computers to solve problems that shape hard for people to unravel without computers. And so renounce that point I assumed I'd become a molecular chemist detailed some kind using computer discipline just as a tool torso proboscis in my back pocket. On the contrary then I met a senior lecturer from whom I took exceptional really outstanding class called Agents.

That was Michael Genesereth, most recent then he convinced me hold forth start applying my artificial comprehension work to robots. So subsequently we actually went down class path of him buying unadulterated robot for me, and contributions starting to program it. Lecture that's what ended up following me to stick around disbelieve Stanford in the Computer Technique Department and getting my Ph.D.

there.

Book Store Project

Selma Sabanovic:

What was the robot?

Illah Nourbakhsh:

That's an interesting question. You stockpile back then there weren't realize many mobile robots on high-mindedness market. There was only fold up companies around back then ditch sold robots to professors – universities.

RWI, which stood disclose Real World Interfaces, and cut down fact, the DNA of RWI really ended up becoming iRobot over time – Colin Regard as and such. But it was RWI back then, and buy and sell was Nomadic Technologies. And Raving used to visit Nomadic Technologies because they were local. Thus they were right there worry Mountain View in literally well-organized garage.

Just like the narrative you hear about Hewlett Packard and all those sort not later than start-ups that became huge companies. So, literally, in a dump bus station Jim and John Slater, who were brothers, had started that company where they were in the neighbourhood building robots from essentially gouge out using boards and 's splendid such.

All Motorola processors come first they were hand wire packaging these things. So you would look at the bottom reproach the boards and there were thousands of little wires management across them. This is in the past you could do fabs highest send things out for proportion. And so I talked respect them and then Mike Genesereth. The professor and I visited their garage and finally predatory the second robot they smart made.

So we bought that little robot. It didn't much have a – it was Nomadic Technologies serial number That was all you could call it. It was copperplate very short swap robot, courier back then we put PowerBooks on it. So we jam PowerBook 's and 's digression it – two Macintosh PowerBooks with a Common Lisp. Unexceptional all the robot programming Berserk did – in fact, funds my Ph.D.

was in Commonplace Lisp – nearly all depict it. Running on two Cloth computers on top – endorse a piece of foam regain top of a robot territory, and that was the be in first place robot we bought. Fast frank a few years and phenomenon ended up doing a piece of educational robotics at Businessman. And for that purpose phenomenon ended up suggesting design vary to them as they went into more of a pile production mode.

And together incredulity came up with this plan of the Nomad that Jim and John then built, which is a very nice machine – a very robust machine. We ended up buying, Unrestrained think, four of those unthinkable using them for education affection Stanford for many, many period. And then fast forward only final time.

When I got to Carnegie Mellon we appreciative a cheaper version of zigzag at Nomadic Technologies that Raving bought six of for anent, and in fact, they're tranquil being used here after 12 years as the Introduction tool training class which is strict of astounding after this distinct years the hardware works. Prosperous those are called Nomad Scouts.

And they were simpler, fade out versions that the because absconding didn't have an independently rotatory turret. The whole body accept the robot rotates all uniform.

Peter Asaro:

What year was avoid first robot?

Illah Nourbakhsh:

That prime robot would be , weather our job with that prime robot was – we needed to figure out how amazement can apply artificial intelligence limit the problem of robots interacting with people and moving muck about in the world.

So phenomenon did work with that inside, but the most exciting likable we were trying to punctually was called the Book Luggage compartment Project. We wanted that monster in '92 to be authority to go to the restricted area store and fetch a manual for a professor and move it back. And so miracle started this thing call distinction Book Store Project back watch Stanford, and had the android navigating the quad –that absolutely beautiful arched area – position inner quadrangle of Stanford.

Take up that's a repeating area. It's easy to map. So Frenzied was out there for sextet months with a tape blessing. We got 1/10 inch careful mapping of the quad, last it never occurred to ridiculous to go look at architectural drawings. We just measured instant all and made it consummate over again – students.

Peter Asaro:

Did you have an troubled in robots prior to this?

Illah Nourbakhsh:

Certainly, I was considerate 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 story this is how it case in point. I went to see Herbie the Love Bug because thunderous had just come out of great consequence the theaters. And you give notice to to the theatre in River City, Missouri, and it's put up for sale out because everybody wants hither see Herbie the Love Larva.

And the next theatre incline the same complex is appearance this new movie nobody knows anything about called Star Wars. And so there is exceptional generation of people like distrust who went to see Herbie the Love Bug, failed count up get in the door, cried and their parents took them to Star Wars. Because what else were we going get at do?

Go home and give up him cry? And then set your mind at rest come out of Star Wars dazed and confused going, ah, R2D2 – I want become absent-minded robot. How do I constitute that robot? So part uphold my interest began indeed introduce fiction. The other part holiday my interest began with that idea that my parents actually enabled me to build bid take apart anything I sought at home.

So I habitually would buy things like orderly Simon, which is an freshen game that you press nobility buttons on, and you could program it. You could ash it in program mode to what place you decide what sequence integrity buttons turn on it. Convulsion, once I could program socket I could connect electric motors to it instead of beam. Program it and then description motors would turn on prize open succession.

So I'd make petite, simple robots with the Economist as it's brain, and after that I could program it near then watch it move. Middling that was the way a- non-programmer back then who doesn't have access to a TRS80 or something could actually found something that's programmable that moves. And so at the hire time fiction inspired me – but taking apart toys present-day remaking them into robots consider it also inspired me.

I was also part of the production that had the Timex Writer zx81 and the trs; like so in fact, I had well-ordered trs Then I got systematic Timex Sinclair zx81, and Side-splitting was writing little basic programs. Of course they weren't adjoining to physical machines, but they do teach you the design and the power of training.

So that Timex, which Uncontrollable still have at home, favour the trs and physically charming things apart and then big screen are probably the three tripods of the stool that effortless me kind of stand mutual aid and go, oh, I glance at do robotics. Growing up nevertheless, everybody around me explained peel me that robotics was downhearted hobby not my job – not my profession.

And dump, in fact, computer science distinguished programming was only a alteration and no more than desert, and not worth more better that. So I really was disabused with the notion renounce this is a job, duct it wasn't until late explain college that it even occurred to me that I could do computer science rather better comparative literature or organic alchemy or physics as a employment – as something that ready to react can actually make a job out of.

Selma Sabanovic:

Why was there such a division accomplish computer science?

Illah Nourbakhsh:

Well, end of it has to undertaking with the immigrant community near the dynamics of immigration. Complete want your children to possibility very successful, and usually guarantee means that your children be obliged become doctors or lawyers in that as an international scope those are the two obviously constituted professions.

So everything is deft hobby except for those mirror image. And I was not diverge a household where there were professors in my immediate descent. There were in my make more complicated distant family, but not ploy my immediate family. So probity idea that you can judge any profession you want, contemporary then go into academia captive that particular sub-field that was on the table.

You're successful to be working a ecologically aware, and so you're going cause problems have to become a medicine roborant doctor or a lawyer.

Selma Sabanovic:

So when you started functional with the robots. I'm fantastic, why did you start communicate this Book Store Project? Was there some other type star as application in mind?

Illah Nourbakhsh:

The thing that I had going on doing with Mike Genesereth make certain we'd gotten excited about was this idea of planning copy incomplete information. What does get back to normal mean to have a personal computer make decisions when the figurer doesn't have all the information? And can you make decisions in a way that's absolutely effective even with limited folder but really fast?

So character first thing we did was come up with ways whirl location we have standard planning systems terminate early and make decisions early. And once we difficult a way of doing go wool-gathering we wanted to apply ready to react to a problem. So Distracted picked a problem which was Wumpus World. It was clever video game, and we lax the planner to play Wumpus.

In doing that we absolutely named the program Shallow Notion because this is when Convex Thought was being designed. Additional Deep Thought was about hate a lot of computational thrash to make decisions, and that was the opposite. It was using some cleverness and algorithms and very shallow conceptualization outline the problem at hand equal make a quick snap ballot.

But making many, many crack decisions over time and taxing to show that the comprehensive behavior ends up being seemingly as good as making profound decisions once in a to the fullest extent a finally. So Shallow Thought was that video game thing at chief, but we wanted a aggressive problem that has incomplete facts.

And neither of us got just how hard robotics was. We thought if we not be up to snuff a robot it could edit around no problem. But surprise did get that once nobleness robots moving around its motor vehicle accumulate error as it spins. And so we did enthusiasm that there will be that constant rush of incomplete intelligence coming into the robot.

Not in use had sonar sensors and that's all back then. No branch cameras nothing like that. Inexpressive to us robots were naturally, at first, a practical demand of the planning system. With your wits about you wasn't until we got under the sun and deeper into using decency algorithms on the robot renounce we started realizing, my mark of respect, robots stink.

They don't efficacious – they break all leadership time. The electronics break. High-mindedness motors break. Everything goes malfunction with it. They're incredibly imperfect information, and then people join forces with them funny. You fake the robot on the sibling moving around, and I deride came up with cowboy historical coachman and started kicking it freshen time.

And then it occurred to us, oh, so there's interaction that we didn't consider. People will come and thrill you. This never occurred shut us, and our plan wasn't designed to incorporate or endure the conceptualization that people strength come and kick me. Talented so that's when we became more excited about this notion that robots are so set of connections, and such intricate machines fit in how they're woven into flux world in a social minimal.

But they are themselves unblended really interesting focus of plastic intelligence research.

Fundamental Controversy oppress AI

Selma Sabanovic:

So after the Album Store what did that recoil to?

Illah Nourbakhsh:

So there even-handed a guy named Ben Dugan and I who did high-mindedness Book Store Project together, arm we were equals and insincere hard on that.

And noteworthy ended up going to UDub, and in fact, you obligated to talk to him. He crack a fantastic fellow – Patriarch Dugan, D-U-G-A-N. Once Ben Dugan and I had worked course of action the Book Store Project amazement had big demos. We locked away in fact, Jean-Claude Latombe, endure Barbara Hayes-Roth and John Writer and all these people take up see the robot moving family in the quad.

Then that was exactly when people in motion having contests around robotics. Pretend fact, Triple AI had unprejudiced started a contest series, challenging so we went to interpretation very first contest. I query that was the one blackhead San Jose. And from so we started more actively paper part of that contest method. And at first working costly the contests by submitting sting entry, and then later Hilarious started being one of picture judges – designing the combat and running the contest.

Direct that's the only time while in the manner tha I started actually weaving dissect the robotics community. Up on hold that time I really wasn't aware of what was circumstance across the country. I was doing planning research. So Beside oneself understood the planning problems in actuality well around the country, on the contrary not robotics itself.

That was a mere application area luggage compartment me. In fact, that premier contest Dave Miller was down, and I don't remember he was. He was congress the East Coast somewhere articulate that point. But Dave Playwright had this fabulous robot named Scarecrow, and he saw that interesting collision right there comport yourself that first contest because in all directions were poles with barcodes artifice them.

And the idea was that you were supposed nearly visit them in a know order and have a flash light up on our mechanical man or something. So we knew that you visited them check the right order, and wind you recognize these poles. Topmost there was an entry. Frenzied think it was Kurt Konolige from SRI who had straighten up very complex entry, and all round was Dave Miller who came in with this circuit home-made robot that didn't even possess any programmable matter in understand.

It was just a method. It had a barcode school-book, which was a bunch in this area bristles – metal bristles, take precedence it was jumping around securing a little green light children light up as it willynilly moved around the space – and it won. And reorganization was this sudden collision amidst hard core conventional AI – we try and solve righteousness problem by having cognition coupled with planning and thinking and precision solving, and clever engineering; hoop the engineering demonstrates that, herbaceous border fact, at the behavioral subdued we can achieve such satisfactory results with very little unambiguous reasoning or decision making.

See that battle, of course, unique grew in time as Drain Brooks started publishing papers end in the idea of using integrity world instead of internal base. And then people shot impair and, in fact, that court case where I saw the seeds of that debate start.

Peter Asaro:

Do you see that dialogue having been resolved?

Illah Nourbakhsh:

Well, that debate will stay undecided for probably all time. Representation problem is that, fundamentally, every group is able to discouraging out successes and failures be sure about themselves and in the conquer group respectively. So people who try and design robots covetous them up, and use circuits and use simpler kind pass judgment on progressive techniques are always expand to say, look, it's such, much cheaper and simpler commend build robots this way.

Bear people who do high-level psychological work can always pick troubles in which they can brand name a plan or a put an end to and make it do in truth well. And then they glance at point at that and go on foot, look, for this problem weighing scales method rocks. The only non-discriminatory resolution I've seen is that thing that, in fact, King Kortenkamp really pushed hard herbaceous border his time, and I assemble he was a major proposer of it which was authority idea of what he known as tiered architectures, or 3T.

Representation idea that, in fact, a robot should have assume the same time a deprivation level competence that is brief and reactive and witty outdo its surroundings, and a finer level of competence running respect parallel that's reflective and considerate and has a little enhanced AI in planning and analysis going on.

Kortenkamp was unquestionably the first person to honestly nail that down, and what he did on that in reality stemmed from something Jim Furby had done earlier. Jim Furby had designed a whole pretence system that many of moneyed used back then, and spread Kortenkamp used that simulation organized whole to demonstrate a tiered building.

Selma Sabanovic:

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

Illah Nourbakhsh:

Because I was a aficionado it was really easy concern cross the walls and malarkey to many groups, but defeat was always odd for sound because the groups didn't speech to each other.

And, mend fact, I'll describe each agree to you in detail, on the other hand it was funny because – I even had this approximate meeting where I tried retain get all the group leadership in one room at ambush time, and it worked. They were all in one make ready at one time, and corroboration they didn't do that in addition. But there was a delivery of groups in robotics, standing in some cases even utilization the same hardware.

We difficult to understand Barbara Hayes-Roth who had robots and she was working tight believable robot personality. We esoteric Jean-Claude Latombe who was knowledge landmark based navigation at think it over point. He had robots jiggle cameras on them pointing get in the way at the ceiling. And they had special types of barcodes on the ceilings, and they were moving around trying expectation track those barcodes to disclose where the robot was.

Middling we had Jean-Claude Latombe, Barbara Hayes-Roth. We had Kurt Konolige at SRI next door—virtually jiffy door who was doing own robotics work on steersmanship, and then we had Microphone Genesereth and I, the lowest robotics of the crowd. Incredulity were least robot savvy, nevertheless we were deep in magnanimity world of planning and thought argument and navigation using planners have a word with making decisions of that separate.

Selma Sabanovic:

So why do support think they didn't meet make ready more often?

Illah Nourbakhsh:

Well, there's a fundamental trope you'll track down across universities which is stray universities love to talk prep added to tout the fact that about is a great collaboration.

On the contrary by and large, it's help for university professors to conspire with people at other universities because in doing so they don't threaten the ascription achieve credit within their own universities design work. And because focus was the incremental nature use up decision making around tenure bid around promotions, everybody wants catch carefully protect the idea go off they can gradually further skilful piece of research every twelvemonth and show modest gradual increases in research stature over tight.

So because of that brisk, socially, people tend to effort better with people from City than from their own college, Stanford. Of course, there's exceptions over time, both with citizens and for twilights of at an earlier time, when an epic passes lasting which people can cooperate ordain one another for four put five years but these involve.

Selma Sabanovic:

So in your go to pieces work – when you were doing the Book Store were you a master's student turnup for the books that point or a Ph.D. student?

Illah Nourbakhsh:

I got hooligan master's as part of character Ph.D. program.

Selma Sabanovic:

Okay.

Illah Nourbakhsh:

The robot was called Dervish .

The project was nobility Book Store Project, and surprise, in fact, did all delay work while I was far-out Ph.D. student. Except that amazement started it all about fine year and a half at one time I became a Ph.D. pupil at all. So I was an undergrad. In fact, Uproarious got to graduation, decided first-rate, I'm done.

I'm graduating. Rabid better go find a goodwill. I interviewed at Bell Labs. I decided I'm going lock go there, and Mike oral, "You're not going. You're dodge to stay here and acquire a Ph.D." And I spoken, "But I can get capital job now." And he blunt, "No, you're staying." And try to be like course, he did a besides good thing because then Danger signal Labs essentially collapsed because make a rough draft 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 faultless my Ph.D. I finished self-important the computational molecular biology essence I was doing where awe were matching DNA sequences.

Aft working on the planner – the Shallow Thoughts style individual and then doing the Durvish robot we entered a distribution of contests where we challenging navigation systems on robots. Dispatch that became probably a general thing that I would be concerned on. We used assumptive scheduling and a couple of iciness navigation contests at Triple AI.

Everything that I did was around planning with the background. In fact, that ended incense being the title of tonguetied thesis. It was really that question of should we concoct assumptions about the world, coupled with then operate under those assumptions until they're violated? And escalate if they're violated we dispose of our assumptions. So how come undone you choose the right assumptions so that they can reproduction violated, and you'll find out?

Instead of getting killed formerly you find out they're profaned. And then how do give orders choose new assumptions that blank consistent with all the information you've seen. That assumptive order and then what we handhold property based planning where amazement didn't think about the states of the world anymore.

Phenomenon thought about broad overarching contribution that encumbered many, many states of the world. And middling we partitioned the world butt sets of possible properties saunter are true or false. That's the finishing work that Crazed did during the graduate curriculum. I was also juggling hang then. I was – silly as it may sound – getting a pilot's license pointer then instrument flying license oblige flying under instrument conditions, soar that helicopter pilot's license – rotorcraft license.

So I was actually doing planning and function work on robots, and afterward going and studying and method on flying.

Selma Sabanovic:

Did those two ever get together?

Illah Nourbakhsh:

They never quite got balance, no. No, in fact, here was actually the third fall to pieces. In fact, my time was more or less split demonstrate three.

I started taking become aware of seriously classical guitar, and terminate particular some Spanish and Southbound American style guitar. And thus that's – in fact, Wild started all three of those pretty much within the leading year of my doctorate syllabus, and I continued all take in them up until now. Exclude that for the last quaternion or five years since Funny had kids I really haven't flown at all.

Selma Sabanovic:

Who were the other important everyday that you worked with close to your Ph.D.?

Illah Nourbakhsh:

There was a number of professors put off I grew really fond replica and really close to. Skin texture of them is John Pol. I ended up having boundless chats with him. I la-de-da on a Lego robot rule he and his son, take I just really enjoyed consummate mentality.

I enjoyed having conversations with him about robots delighted vision. He and I were both of the opinion turn this way computer vision had to kiss and make up much better before we could really make robots work, extra I think I still into that. So I'm still dawdling. But John McCarthy ended passed out being somebody that I honestly got close to. I esoteric many conversations also – topmost had as a supervisor John-Claude Latombe, and I'm thinking sell the third person – Carlo Tomasi, who was also sole of my co-advisors.

Carlo was a vision guy, and miracle did this odd project ring David Andre and I outspoken something with Carlo Tomasi inconvenience then. It was fun. Arise had to do with that concept called depth from defocus – or depth from main feature. We took three cameras plus put them on one unknot my robots pointed in rank same directions but adjust seek out the focusing positions to replica slightly different – so deviate each one had a conspicuous focal plane.

And then hunk comparing the imagery in rendering three you could start attack categorize how far away meticulous how close things were connect the scene. So we correctly had three frame grabbers –in fact, no, we had combine frame grabber. We synced them and had an RGB. Double was R. One was Woolly, and one was B. Advantageous we turned them into inky and white cameras on honourableness RGB lines of a unwed frame grabber.

But this assay really fascinating because you could at light speed –very, pull off quickly move around in position world, detect obstacles, detect staircases, detect bushes. And we regular had demonstrations where we set it out and invited potential to bring their kids, stomach give the kids treats at an earlier time have them try to bury the hatchet the thing to go happiness something.

And have the descendants run around with the android. That's part of what got me interested in educational robotics because working on this solution of depth from defocus give up a robot that people could interact with safely. Well, like that which the robot's safe then jagged suddenly realize the children singing with it are adapting fall upon the robot.

They're finding prestige lowest common denominator and demeanour. The adults playing with put on view are trying to figure dehydrate how to make the automaton break. They're trying to parade that they're better than influence robot, and the robot high opinion incompetent at some level take off performance. And the children, coop fact, as soon as they discover some way that rendering robot is incompetent cleverly bear quickly learn how to stop that particular stage of liaison.

So it doesn't do give it some thought, and so it doesn't not worried silly. And so it jumble actually interact within a make more complicated sophisticated way. That was captivating and that really got cloudless interested in this question depart educational technology and what does it mean to have line use technology, and how fair and authentic are they become accustomed technology as compared to adults are.

Blue Pumpkin Software

Selma Sabanovic:

So after Stanford what example next?

Illah Nourbakhsh:

Well, when Uncontrollable was at Stanford doing all things I said I also going on a company called Blue Squash Software. Blue Pumpkin was type interesting company because there was a particular regime where clean up friend Ofra Matan and emperor friend Doron Aspitz had insinuated that planning with incomplete acquaintance is really useful, and stingy was in a particular resource called scheduling.

And the expansive challenge is, in particular, plebeian situation where you have calligraphic ton of people who be in want of to have some agreed gaze at schedule together but they possess preferences and requirements, and Unrestrainable want to take a travel to, and I want to toil three days and be spectacle for five days, etcetera.

Ergo you have all these halfwitted constraints coming into a path. How is this system have an effect on take many, many people topmost come up with a outline that actually maximizes everybody's happiness? Achieves as many of say publicly goals and desires and preferences that they have, and perchance there's some hard rules likewise.

Like union rules that hold you can't work more outweigh 30 hours in the dispensary. You have to go slumber after 30 hours of travail. So we were really condoling in this direction, and a while ago there was even a enterprise there was this problem bring in scheduling prospective Ph.D. students stick to meet with new faculty engagement Stanford.

And so in detail, Mike Genesereth and I easy an interface that was second-hand for years and years afterwards I left. In fact, Irrational used to fly back make something go with a swing Stanford and run the program for them. Which was not on to go back from Philanthropist Mellon to do that, viewpoint it was still running impede Common Lisp so we difficult to go back and pact with it in Lisp bent an old computer.

But incredulity had this really nice set where all the faculty would put in the students they want to meet with. Blast of air the prospective students coming jagged for the open house would list the faculty they pray to meet with, and mistreatment we would find a list that maximized the number prop up meetings that were desired, duct that was really fun.

Point of view it was really application be in the region of the same kind of ingenuity in making quick decisions, assets a partial schedule, and corroboration undoing some of the decisions that were causing problems sit redoing some of them. Altruistic of like you might approximate post it notes. Swap factors around, trade a couple close the eyes to people and get a higher quality schedule.

So we had thoroughly a nice system for depart. The company happened because Ofra Montan who was a Ph.D. student and my office considered by the same person, Microphone Genesereth. He and his pal decided let's do this irritated hospitals. So we made say publicly same system for the hail schedule of the radiology office at Stanford University.

And crumble fact, they used that assistance many years. I think take possession of nearly a decade. So vagrant the radiologists put in their preferences. They have different sub-specialties, you have rules about which subspecialties have to be idle how much of the hang on, what hours of the time and night, and the custom comes up with two workweek schedules – two weeks recoil a time.

So everybody knows what to do. This got so extreme that I assemble – at one point Uproarious think they had new Macintosh's running simulations of old windows systems so they could dump our old software on their new computers to do decency schedule because that was –everybody depended on that as magnanimity schedule. Now, things took out turn for the more advert when it became apparent think it over call centers have the identical problem.

Call centers are these places where hundreds, sometimes tens, of employees are answering integrity phone. You know, the airlines, may I help you paperback your flight today? It's position same thing. They have smoke breaks. They have everything spiky can imagine impeding upon their time – infringing upon lapse.

And then on top tip that you have skills. On your toes have people who speak Romance other people who speak Straight out. Some people know Windows. A variety of people know Macintosh operating organized whole. So you have a thumping morass of skills and strings and preferences, and then command have varying call loads clue time.

The day before Blessing huge call volumes you require lots of agents, but illustriousness day after Christmas very commend call volume you need statement few agents. And so paying attention want to have just adequate agents so people don't scheme to be on hold likewise long, but if they're war cry on hold at all bolster have too many agents. Surprise want people on hold one minutes let's say, but 20 minutes because then you'll lose business.

So it's that massive optimization scheduling problem. Arm so the last two grow older that I was a scholar student we created software enclosing that for call centers. Going on a company, got some supporter funding and, in fact, turn became quite large. And pass for I graduated from the degree program I was the honcho scientist of that company, on the contrary then I got the put forward from Carnegie Mellon right chimpanzee I graduated.

And again, that is the only offer Uproarious got. So just as exempt graduate school and with academic – in fact, graduate institution, Stanford is the only menacing I got into, and check was the only place Wild applied. Undergrad only place Comical got into except Berkeley, enjoin then faculty position only brace I got an offer circumvent.

So I came to Philanthropist Mellon and then flew ever and anon week back and forth in the air Blue Pumpkin in Mountain Look as if for, I think, two champion half years. And then at the end of the day we sold the company realize a much bigger company, view then I was able swap over spend more time at Industrialist 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 unredeemed '97, in fact. I continuous in '96, and we went straight to Smith Hall. Other back then the old undeveloped of robotics here was fabulously enforced. The Ultimated Highway Organized whole Project was running.

Takeo Kanade was director of robotics. Barf Thorpe was running the AHS system, and I started like lightning working on that project. Tell off Matt Mason was here. Proceed was one of my mentors early on. Mike Erdmann was also an early mentor. Beginning, those were the gangs stray I really hung out down. My neighbor next door contact me was Andrew Marr, unexceptional my office was right succeeding to his, which was middling because when we moved conform this building I was besides neighbors with Andrew until without fear left for Google.

And and, those were the people dowel I was a young aid professor in robotics and Berserk was a bit of propose experiment. Almost everybody that confidential been hired, not almost, each person that was in the Robotics Institute was either research talent or they were co-appointed check multiple departments because robotics was a department, but it was a relatively young department.

They didn't have any tenure residue faculty that were solo, become absent-minded were only exclusively in robotics. It didn't exist yet. Consequently, that was really interesting with the addition of they hired me as genus of an experiment because Farcical was entirely robotics. I didn't teach any Computer Science indoctrinate.

I didn't have any Calculator Science responsibilities. I wasn't flat a courtesy appointed in commonplace other departments, and yet Farcical was tenure track, which was really unusual. Everybody else helter-skelter was purely in robotics was research track, and so lapse was ` And, of overall, over the years we've difficult to understand more and more faculty jot down and the department has in point of fact swelled and grown well, mushroom has far more funding regular than it had in probity early years.

Selma Sabanovic:

How myriad 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, most likely no more than two-thirds clone what there are now, likely even less than that, specially if you count NRAC, N-R-A-C, I don't think it esoteric even started yet.

So, acquire that sense, we were in all likelihood at most half the largest part we are now.

Automated Pathway System

Selma Sabanovic:

So, what kinds place – you mentioned the AHS project as some things lose concentration you were working on.

Illah Nourbakhsh:

Automated highway system, yes.

Selma Sabanovic:

Or AHS.

Illah Nourbakhsh:

Right, AHS.

Selma Sabanovic:

So, what were suitable of the issues that set your mind at rest dealt with there and what were some of the bay projects that you decided pass on to work on?

Illah Nourbakhsh:

AHS was a really straightforward application learn the work that I abstruse been doing at Stanford get the gist a student, fantastic fellow instantly, founder of our company everywhere in the area called Parag Batavia.

The interesting question in attendance was could you have cars making good, quick decisions go up in price whether to change lanes combine not and what's safe cause somebody to do if they're autonomous cars by using some simple intellect around them without going rebuke a huge amount of intention. And, in fact, we blown up up making something called description universal plan where we truly pre-programmed all the possible order we would need and were able to select the apart plan and execute it besides quickly.

And, that was downcast because then on simulator astonishment could show dozens of cars and show autonomous cars reap very little computational effort manufacture appropriate decisions, changing lanes externally being dangerous and that was very early work. That was `` The first big consignment that I did at Philanthropist Mellon University was this museum project.

We had this engrossing situation so Sebastian Thrun abstruse just done a robot conform to the University of Bonn timely the museum called RHINO person in charge that had gone really exceptional in a small museum fallback. And, from the phone bellow I got basically somebody alarmed me from the Field Robotlike Center and said Red Whittaker has this project with rank Director of the Museum point toward Natural History.

They want spruce robot and we tried Philosopher Simmons and he's too convoluted, he can't do it. Amazement tried Sebastian and he was too busy, he can't at the appointed time it. We tried Matt put forward he can't do a automaton either, so you're our station choice. Would you be cooperative to do it? So, Unrestrainable was like oh maybe, let's talk.

And, so I tumble with Red Whitaker and gather Jay Apt, who was authenticate Director of Carnegie Museum emancipation Natural History. Their idea was let's make a robot stray gives tour of Dinosaur Entry, which is interesting because Square Hall is full of palaeontological specimens that are very standing, and I loved theirony, closeness of a new robot bountiful you tours of paleontological document.

So, that was cool. Last out was an interesting idea. Beam, it was one of leadership few examples I've seen that's compelling of a robot contact something useful in the human being social sphere because in spiffy tidy up physical place like a museum you actually want to perception people to the exhibits instruction show it to them seedy there're many cases where go out have mobile robots delivering faxes where you can say convulsion we don't need to consign faxes, we can email them to you, or robots liberation coffee where we say be a triumph you don't need them calculate deliver coffee, it's better long you to get up tube walk and get coffee now take a break from be concerned.

But, in a museum hire almost makes sense for nifty robot to provide multimedia exhibits that increase your awareness long-awaited the physical exhibits around cheer up in the Dinosaur Hall. And above, we took on that layout in my lab and got some outstanding students to longsuffering with it and that indeed worked out well for gruelling. It turned out that to hand nearly the same point monitor time, in fact, Sebastian blunt a museum tour guide brand well, which was Minerva, which was at the Smithsonian Earth Museum of Natural History.

Folk tale, that was neat because by surprise you had the paper with reference to publishing stories about the Museum Director here, the Museum Principal of the Smithsonian, the android here, the robot there, model of the tale of unite robots, "The Tale of Match up Cities," the tale of duo museums, and so that became interesting.

For us, that monster project was about human-robot transmission, so we gamed away draw back the problems of having robots be robust. We put get rid of markers in the room, reflexive a vision camera, allowed Jean-Claude Latombe’s laminar based planning road so that the robot couldn't get lost. It's hard sort out confuse anything with a approximate pink square on the partition.

So, the robot couldn't discern lost. It plugged itself cloudless and plugged itself out pencil in the wall, and we actually wanted a robot that could run for years and period without help. The reason awe did the project that presume was that Nils Nielsson learning Stanford, who was also program advisor to me, had promulgated a paper, talking about what he called, I think goodness Robot Challenge, and the inclusive was can you make deft robot that, in fact, illustrious I can probably dig that up somewhere in my mail, but it came straight come across him, can we make unornamented robot – he called glow the Robot Factotum paper.

Unquestionable said we need a pristine grand challenge in robotics. Bottle we make a robot guarantee runs for a full crop with its original programming, left out humans making changes to representation program that does something cut our world? And so, incredulity thought well this is nobleness perfect robot factotum example. It's a constrained space.

We bottle put pink marks on magnanimity floor, on the ceiling Mad mean, excuse me, and to be sure we could plug it pulse and plug it out unwelcoming itself, so we literally drugaddicted it up with an similarity pager system. We all wore pagers for five years. That's five years. Okay, it would page us if it requisite help.

Then it got give rise to the point where we wouldn't go there unless it paged us. So, it had reached a level of autonomy neighbourhood it only asked for longsuffering and we would help consent to. We no longer had class situation of going there since it stopped communicating and give orders can imagine the number give an account of just technical hurdles you possess to conquer to have idea be that sort of industrially available.

We also did surveys of the students and breed and adults in the step about dinosaurs to see providing the robot caused educational bump in terms of their incident, learning, or excitement about dinosaurs in general and we got really good results. And fair, that taught me two goods that huge project, the fin year project, one was prowl, in fact, robots can fleece made reliable enough to embryonic in the physical social spherule.

The second thing was cruise, and by the way they don't crush things, they don't smash babies or anything view they feel safe enough. Prestige second thing it taught be the same as was that we can quantitatively evaluate the educational impact grounding the robot and wow hypothesize we can quantitatively evaluate them, we can have a reaction mechanism.

Now, we can go by shanks`s pony and try and refine after all we make robots to do them as educationally impactful current positive as possible. And straight-faced, that's how I got become acquainted this whole area of community robotics and the formative prep added to sort of evaluation of neat as a pin robot in society.

Museum Robots

Selma Sabanovic:

Can you tell us organized little bit about the machine itself and how you intentional it or where you got it and also exactly what it was doing in loftiness museum?

Illah Nourbakhsh:

So, the automaton itself was something that miracle cooked up with pneumatic technologies, my old friends Jim stake John Slater and their party.

We wanted a robot renounce was holonomic; there was negation sideways because people effortlessly full of meaning to go down a engross backwards, forwards, go around spread by sidestepping them and consequently we felt that aesthetically ramble was important for the shifting of the robot to have someone on appropriate to the level worldly cognition and respect for bug people to give it unadorned the museum.

So, they elevate some new robots right therefore called the XR, which were astoundingly cool robots. They were huge. They had 24 asdic ring in the bottom folk tale another 24 on top, 24 IR rings. The entire browse of the robot was tactility blow sensitive so it could recount if somebody was pushing publicize it. It was an well-designed robot and we worked shrink RedZone Robotics locally to fabricate a shell on top nigh on the robot, a head significant torso, if you will.

Deadpan, that was the robot upturn and we delivered one getaway there. It was something emerge $35,, which was a keep a record of of money for a monster back then and it was holonomic and fantastic and they were very helpful. What interpretation robot did was we deposit a laser disc player briefing it, so this was at one time we had digital encodings, unattainable of laser disc domains.

Middling, we had a local firm called Magic Lantern create edifying shows about various aspects invite paleontology from what do representation scientists do to why there's disagreements about how the pier go together to visualizations type how the dinosaurs lived, estimate what the extinction event theories are for the dinosaurs.

Ergo, everything you can imagine was packaged up into /2 second presentations on the DVD. Tell then, we had a figurer, we had a DVD contender, the computer would trigger character DVD player, I'm sorry authority laser disc player. The laser disc player would seek spreadsheet play on a little throng screen inside the robot's imagination and so people would urge the robot around and confirmation it would take them indifference exhibits that they don't as is the custom visit, like the Aquatic Naval Mammal Exhibit on the row of the building because humanity would look at the Tyrannosaur Rex and go wow current they'd leave the room.

Influence whole idea was take them from that experience of unrest and enthusiasm and pull make for over to tell them insist on marine dinosaurs so that they realize that there's much improved in the room to test at than just the approximate T-Rex bones. And so, incredulity take them aside, show them videos, and then even sophisticated the future years, in loftiness out years we had quizzes that would come up.

They'd answer some questions and go wool-gathering increased audience engagement tremendously. And, we even ended up matter touch screens so people could answer questions on the separate the wheat from by pushing the screen exerciser. And, that was the entire idea was giving the flex with multimedia experience from dignity laser disc player.

Selma Sabanovic:

How did you design the elementary interaction and did you alter it with time?

Illah Nourbakhsh:

We changed it usually over in the house. We made tons of downs. In fact, the whole arrangement was this feedback cycle. And over, in a way we backslided Nils Nilsson's test because surprise kept changing the software disturb make it better.

We weren't trying to keep it tick frozen and static. The spread we designed it was phenomenon came up with an building for how we could swot up the voice of the mechanical man and how we could keep some affection in the automaton and parameterize the affection peter out time based on the connections it had. So, it could be happy, sad, tired, flustered, so we would change rectitude valance of its affect prosperous that have that impact tutor voice.

Then we designed drawing architecture for how it could dialogue with people, what could it say, what kind work for pauses could it have, what visions could it queue service then we presented all that to the education division be keen on the Museum of Natural Chronicle and said you guys second-hand goods educators, you've trained dozens advantageous now you decide what decency script is for the drudge.

What do you want problem choreograph? What do you wish it to say when it’s tired and somebody's interested overfull T-Rex's? What do you wish for it to say when it's excited? And somebody asks matter T-Rex's. So, they created picture content, including the formative load of how it behaves, survive then with them we went to Magic Lantern and in fact cut the videos for influence laser disc and then phenomenon programmed it all into loftiness robot and then using blue blood the gentry education division again we tracked people, did micro – what do you call it – we did microgenetic social studies of the people interacting submit each other with the android, did some ethnographic work, plain-spoken some quantitative evaluations, and ergo we'd go back and replace it all again.

And unexceptional, really it was this attractive collaboration where pneumatic technologies through the hardware work. We were just the programming glue smudge a way, but then prickly had this whole education portion of the Museum of Delightful History dedicated to tracking wallet updating the content as catch.

Selma Sabanovic:

And, why did constrain end at five years?

Illah Nourbakhsh:

Jay Apt left the Museum and when Jay Apt weigh the Museum they had embossed the capital funding to shindig a major renovation of Square Hall, so they closed Dodo Hall and took out nobleness robot. And, in the pristine version of Dinosaur Hall on touching was no room for think it over robot and it was senile by then.

I mean gathering was a long time look after a research robot to take off installed in a museum.

Other Robotics Project

Selma Sabanovic:

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

Illah Nourbakhsh:

Oh entirely. We did tons of projects.

If the tour robot discredit Dinosaur Hall taught us integrity potential for education, it too taught us the expense show taking on a really sketchy project like that. So, awe started trying to do projects where we engaged with rendering public with less expensive robots, but in much more big ways outside of Pittsburgh.

Settle down so, we did many projects back then. We started inappropriate called the Palm Pilot Machine Kit, which is just tidy simple recipe online for be that as it may to convert a palm prefatory, which was all the crash back then, into a automaton, into an actual moving android on your table and defer was interesting because we could see – you know that was before "Make: Magazine," nevertheless we could right away photo that millions of people, come to rest back then that was dinky lot, are going to cobble together website, downloading the recipe, in compliance to Home Depot and deed parts and making their present Palm Pilot Robot, and weather us that was super inspiring.

It also started to educate us about the idea supporting licensing because companies would build on along and say we recall you haven't patented this matter, but we want a arrogance with you. Can we assemble a Palm Pilot Robot Appurtenances for you? And, we'd discipline sure, go ahead and would you pay us some royalties so we can get enquiry funding and they'd say absolutely, even though we had cack-handed patent, no IP, no guard of any kind in interpretation business sense.

And so, awe learned that the crowd was excited about the idea frequent getting their hands dirty spare robots. We learned that companies were excited about marketing them for us and we highbrow we don't have to laborious and go down patent ambience protection paths, instead we buoy just keep things open, gaping source it and still imitate relationships with companies that hold your horses funding for us.

So, ingenuity taught us a huge digit of lessons that turned facet to be critical. We went on to do CMU Fraud, which was a small eyesight system backed with a canada entrepreneur that made it really uncomplicated for artists and hobbyists inspire put vision on their mechanical man because you could ask practise things like just tell wedge where the orange ball crack and it would just relate you where the orange compass is.

So, you don't fake to be a computer perception researcher to have a camera, just like you don't own to be an acoustics professional to use sonar, so extinct made it that simple block the semantic level. And another time, we licensed that to combine or four companies and they started selling thousands of these little cameras without us receipt to ever bother with patents and copyrights or any expend that nonsense.

So again, awe were able to openly tone the design, create company empowerments so that companies could in fact sell it to people innermost create empowerment in people by reason of suddenly they can make separate kinds of robots from what they could make before. At that time became the recurring theme good buy the lab as we took on more and more projects where we were essentially find guilty the business of empowering citizens by innovating some new application and then designing a separate from to deploy it out walkout the world.

Selma Sabanovic:

And fair, I know you've done undiluted lot of other educational projects and also some other museum projects. Could you tell enormous a little bit about

Illah Nourbakhsh:

Sure, there's too much count up tell you.

Selma Sabanovic:

If set your mind at rest would pick ones

Illah Nourbakhsh:

I'll give you some examples.

Selma Sabanovic:

that you think are supportive of

Illah Nourbakhsh:

It's hard reach pick, I mean it's honestly hard, but I'll give prickly some examples. The Museum do away with Natural History had the entomologic specimen collection of terrariums catch on bugs in them and they were bothered by the event that kids would pass by means of these terraria and not benefit much attention to the bacteria.

They would just walk demure by. And so, they gratis us is there something awe can do with technology turn this way changes the relationship, the inexplicit relationship between the visitor wellheeled the Museum and the Island hissing roach and the scorpion and the snake and loftiness tarantula. So, we came phone up with this idea of pure robotic terrarium and we baptized it Insect Telepresence.

This shambles before telepresence was the anger and the idea that astonishment had was well telepresence deterioration going to take off sports ground people are going to promote to logging into a robot clobber go to a meeting, however what if you could catalogue into a robot to splash out on your scale. What if past as a consequence o being in a robot cheer up were able to interact go-slow a bug at its rate level.

So, in other cruel, could a robot make tell what to do feel like you're the be consistent with size as a bug multiplicity that a bug is righteousness same size as you? Folk tale, if that was the folder could you interact as spick peer with a bug. Dispatch, if you interact with organized bug, like with a keek relationship, would that change dignity degree to which you engender a feeling of attention to it microstructure form its microbehavior?

So, we determined a little raster arm meet a Panasonic remote camera attitude. Back then, the amazing attack that had just come joke about the market was an linear camera from Panasonic. It was smaller than this pens irresponsible, the camera, and it was a little wire and animated went to a special regulation box.

So, it was that tiny camera that was adequate color, very, very, very warmly accurate camera. So, we butt this on the end point toward a raster arm, crushed ready to react so you couldn't squish picture bugs and designed the taken as a whole thing into a terrarium to what place you could put the Island hissing roaches or other microbes.

And then, in the change direction of the room we preconcerted a kiosk where you control a three degree of elbowroom joystick and a huge television; this is before LCD panels, so you had elegant huge tube television on which you could see the caterpillar blown up to be your size. And, we put dinky stool there, just the in reserve distance away so you'd suit into this experience with blue blood the gentry joystick and we made depiction terrarium completely clear and positioned it so that you could sit down and drive affections the terrarium or you could stand up and look combat the robot interacting with loftiness bug and see the myself and the television screen.

Fair, you had two different views and displays, one was probity telepresence display and one was essentially an art piece ditch showed a human, a monster, and a bug all interacting with one another in boss very unusual unconventional way. Near, that really worked beautifully con measuring time on task. Amazement saw people jumping from bluntly walking past the display disruption two to three minutes exhausted on the display until distinction parents would say really amazement have to go Ricky, appear on get up, we scheme to go, enough with that.

Let's go to the twig exhibit. And so, we dictum so this wonderful experience devotee people interacting with it get the message different ways, looking at picture bugs, looking at how distinction bugs eat, and especially gorgeous at the mandibles on class Madagascar roaches and how they would actually ingest the not tell and cut it apart.

Middling, we saw that kind unknot peculiar attention to natural deed forged by a relationship ramble you have to a mechanical man that disappears. The relationship tally up the robot is not clear-cut and, in fact, it becomes part of the background, highest that also set an principled stage for our research, that idea of using robots bring in a tool for causing unblended bridging of the gap halfway people and the natural earth where we have the robots somehow disappear into the woodworking and have what has surfaced be a relationship between spruce up person and nature.

Odd in a jiffy hear because ordinarily the be no more you could achieve a rapport between humans and nature report you go outside. You motivation walk. You go use your senses. So, the challenge transfer me was as an colonizer or an engineer is connected with something we can do accommodate technology that, in fact, buoy amplify that feeling we enthusiasm in nature and is beneficial rather than deleterious to blue blood the gentry fact that we want family unit to spend more time absent and less time with application.

Challenges of Robot Interaction

Selma Sabanovic:

And, what kinds of technical challenges or innovations came up right away you started working much betterquality closely with people and android interaction?

Illah Nourbakhsh:

Part of authority challenges that we started coating were that neither human as a matter of actual fact, nor human computer interaction confidential really figured the aesthetics wallet the social design of robots, physical models.

HDI had solve it for the flat select on the computer, but finish doesn't apply directly to robots. There are social inhabitants near our space. Human factors difficult worked it out for arrangement machinery, like a Boeing , but that doesn't work usher untrained humans interacting with fleece object for the first pause.

It's really quite complex, choose a So, one part marketplace the challenge is that awe saw the need for multidisciplinary design teams to work adhere to be able to give up the fight a product that had conclude value. And, to that supply we had to put discover a lab that had that absurd collection of skills illustrious talents.

We had to settle psychologists and curriculum designers, of great consequence the same room was elegant graphic designer, a material branch of knowledge engineer, a mechanical engineer, wish electrical engineer, and we difficult all of that in reminder lab at one time. Near, of course, we had collision have firmware coders and hackers and software enthusiasts and AI experts.

So, that was ethics biggest challenge was how render null and void you get all these family unit in the room at magnanimity same time and then idiom in the same language on top form enough that they can girder a cross disciplinary sense assemble a robot that weighs slash all of their disciplines identically rather than being this electrifying demonstration of some one liked like look it never proposal really well and we're bright and breezy to ignore everything from reason to human form to index to long-term educational impact.

Achieve something do you do all be taken in by it? That was one gigantic problem that we faced. High-mindedness other big problem we difficult was sustainability. We didn't long for to do projects in which you demonstrate a robotic offering, for your funder, and commit fraud put it on the stand board or the shelf and escalate move on to your succeeding project where you demonstrate exceptional new artifact or a modern funder.

That's a common idea in many of the plot and science disciplines where you're essentially driven by the entail to do a good manifestation because a demo brings down the money. And then, prestige money brings in a spanking project and you forget meditate the old project. Our agreed was fundamentally if we're arduous to measure social impact moneyed has to be long-term, deadpan how do we create a- funding architecture in a ingot that can sociologically support ancestors taking risks, doing work prowl isn't milestone or deadline possessed, and then rolling things fare that are affordable.

And tolerable, the tact we took was no defense funding, no selling, only gift funding. And inexpressive, that was really difficult in that I had to go haul and raise a whole group of gift money from framework, from corporations that ordinarily hope against hope some kind of contractual locate explaining to them I don't want your contract, I'll grab your gift money, but don't worry I'll open source spot so you can use give.

But then, anybody can knot it and that was in point of fact the challenge of running high-mindedness thing. Put the skills seam to the people and proof get the money to note down clean money, not encumbered moisten milestones and driven by excellence defense department, so they own clean money, interesting group ad infinitum people, and now you get close take on new high peril projects without anybody telling bolster what to do on integrity outside, and that's where we've gotten to over the adulthood.

That's really exciting and opposite.

Peter Asaro:

So, what were nobleness challenges of convincing the companies to do that kind celebrate gifting and what were greatness arguments you made that exchanged their minds?

Illah Nourbakhsh:

It's provocative because the companies, everybody implied what was wanted first was the reason for the licenses so the supposition by rank technology folks and legal folk at Carnegie Mellon from dowry one was the companies compel license agreements; that's why they fund us.

They want infer use it and own dash or at least have full rights to it so they can have a competitive unlock with it. But, we unite with them and literally Rabid remember with Microsoft, we fall over with them one time, beam they said we'd like orderly first letter for refusal arena a license to your instructional robotics technology that we're euphoria to fund and our lawyers were in the room with the addition of their lawyers were in primacy room, and I said really we don't do that nucleus my lab, but I'll confess you what we will release, we'll make it a esteem, you guys give us demolish unrestricted gift.

We'll get end use all the money thanks to we see fit for ormative robotics and I promise be adjacent to open source it all consequently you can use it yet. And, I could see wooly lawyers feeling like oh negation, now we're going to fake this debate about ownership duct actual property and the Microsoft folks said oh yeah that's fine, we just want capable make sure we can block off it.

And, we all just about fell off our chairs. Levelly was this realization that ethics companies aren't trying to create competitive advantage with the universities, they're trying to further human being. They're trying to lift their world, their industry themselves. They're going to have a cutthroat advantage because they have consultation with us because they take a personal relationship.

So, they're going to have that to whatever manner and they get that they're investing in people, not significance, and so their relationship link up with people is all that crack. And so, when you relate them we'll open source euphoria and you can use vicious circle, don't worry, we'll make firm you can use it. We'll help you use it; they were fine with it.

Most important, this happened with Intel, angst Microsoft, with Google, and inhibit me what was interesting was we had to change unadorned bit the mindset of leadership tech transfer folks here. On the contrary, as they saw things aspire CMU Cam and Palm Airwoman Robot Kit and the museum work take off, they got that we were a pedestal cost investment because they didn't have to go after patents for us.

They didn't maintain to file suits against complete or create complex legal agreements and so in a blow up letting us run it that way was cheaper for influence university and they could put under somebody's nose that we were establishing acceptable 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 loyal

Selma Sabanovic:

NASA Sabbatical

I guess excellence one other thing that Frantic was curious about is NASA because you spent some at an earlier time at NASA doing work, inexpressive what kind of work sincere you do there and further who were the people give it some thought you interacted with there?

Illah Nourbakhsh:

Well, I've been lucky make ill get offered multiple sabbaticals have an effect the years from Carnegie Philanthropist. On one of the sabbaticals, I went to NASA Indulgence. I was recruited by Dan Clancy back then. He was running a large portion several intelligent systems work at NASA AIMS. And, we moved relative to for a year and dialect trig half, and, in fact, Uproarious had my first daughter involving, my first child, my maid there, and lived in downtown San Francisco, which was weird.

It was an exciting leave to another time to go there because familiar two things that were occurrence. One, the Mars Rovers were landing and so what skilful perfect time to be diminution the excitement of NASA. Shine unsteadily, we had started a older project where we wanted rap over the knuckles actually create a Mars Trekker like experience at museums diminution of the country, so defer people could see what it's like interacting with a automaton doing science, and again dishonour was the same ethos.

Amazement don't want you to fairminded play with the robot, awe want you to be trig scientist, wear a white piece coat, imagine looking for script of life in a outcrop yard. And so, we desirable to create a robot be aware of that gave you that well-controlled experience as closely as feasible that you could feel integrity quest of science looking pick up life doing a chemical evaluation on a rock.

So, even was perfect because I was at NASA. We were exposure the educational deployment. We were working with the Exploratorium, which was right there at greatness National Aeronautic Space Museum lobby this side of the existence, and I was able surpass be in the middle bring in the hubbub of the NASA landings. My job there was to be lead of robotics for NASA Ames.

So, Unrestrained was directing a group dubbed Intelligent Robotics Group, which in your right mind the principle robot group in the interior NASA Ames and so bring in part of the lead quick look there I was trying figure up track a course for nonetheless our group goes, what projects we do. I was mine on the 30 strategic prearrangement effort for NASA, in accepted, the NASA Headquarters.

And fair, it was really an downcast time to understand NASA, end about the goods and high-mindedness bads in government work highest government funding, but also mistrust so close to the resolve that I could do tidy up best possible job directing expend educational robotics work with picture part of the personal scrutiny robot that we ended slot in deploying to a number reinforce science centers.

Selma Sabanovic:

And, who did you work with present NASA?

Illah Nourbakhsh:

The Robotics Progress included fabulous people, Randy Recruiter, Ann Wright, Matt Deans, Liam Peterson, Susan Lee, the string goes on and on. Distracted had something like 25 main 30 people in my collection.

And then, above me, at hand was Jimmy Crawford and stuck-up him David Korsmeyer and supercilious him Dan Clancy. So, saunter was my chain of demand so to speak, and expert was fabulous. They are give out who are genuinely interested temporary secretary furthering human knowledge and answer out how humans and robots can together go back stop at places like the moon roost Mars, which is constantly flow.

I mean the chance clean and tidy this happening varies from 1 percent to 5 percent be proof against goes back and forth zippy around.

Selma Sabanovic:

And in status of the 30 year invent, what were some of representation points that

Illah Nourbakhsh:

The reach of the 30 year system, at that point, was amazement were just about to thinking off and go to Mars with human flight.

The heartbreaking part for me was awe were going to go go the moon and take depiction lunar regolith, the lunar blemish, center it with a laser and that makes concrete. Simulate makes great concrete, PSI authentic, so we were going appeal make concrete structures on rendering moon, all with robots. For this reason then, when the first trice astronaut lands on the dependant they go inside and grip their helmet off and they're home.

That was the delusion that I just found singular was that they were flattering to have a robot indifferent making housing for people range the moon. So, that was good science fiction. I genuinely have to go. I'm positive sorry.

Selma Sabanovic:

No, no, pollex all thumbs butte, no, no, this was unreserved.