 My name is Kavi Arya I am faculty in the computer science department and one of the subjects that I teach is embedded systems embedded systems basically are the intelligence if you like embedded in devices and it needs a totally different approach to build these kind of systems as opposed to your conventional software systems like your banking verticals and conventional software like spreadsheets and all these kind of things. So, what I would like to share with you in this short kind of 20 minutes presentation that I have is an interesting project at IIT Bombay which we would like you to participate in in some way called E Yantra. So, what we have done is that over the last 5 years in teaching embedded systems through a distance education program we found it very difficult to motivate students. So, we found that unless students have actually programmed and done a lab where they made things move and lights flash and things like that they do not seem to have learnt very much. So, we came up with this idea of teaching through robotics robot is a very nice instance of an embedded system. So, over the last 5 years we have developed a range of robots in the lab which now we have started spreading through engineering colleges we have covered some 180 engineering colleges all over the country. We have just come back from giving a very exciting workshop in Srinagar in Kashmir and at the moment while we are talking there are there is a workshop going on at IIT Bombay for the colleges of Mumbai university and the vice chancellor himself has insisted that all his engineering colleges be exposed to our workshop. So, what we have done is that 55 engineering colleges and from each engineering college one faculty and two students are attending this workshop and we have a series of five back to back workshops and what we are doing is that we are starting them with the first few steps towards an interest in robotics through the robotics workshop that we have developed and I will show you some examples toward the end of this talk as to what you can do once you attended this workshop. So, a little bit about this E-Anthra project incidentally this is just a small part of the embedded systems course that we teach. As part of the course we give our own students in the M Tech program a two day workshop like this to get them acquainted with the robots on which they will do all their assignments. So, a little bit about this. So, I will talk a bit about robots for college education and for doing much more than that and talk about the present status of our project. But one little thought I would like to share with you it might not be anything about engineering, but it is just to indicate with you that no predictions based on the past are valid nowadays because the world is changing so fast. I often share this statistic with our students that is I heard this at a talk by a leading stockbroker Ramdev Agarwal at a talk he said that it is taken India 60 years to become a trillion dollar economy trillion dollars is when the world starts looking at you differently you become a huge market you become empowered you are a rich man basically and it is taken 60 years that means 2008 India crossed that mark and how long do you think is going to take for two trillion dollars anybody would volunteer in the audience here two trillion dollars how long is it going to take us from 2008 10 years is a possibility actually it is about 6 to 7 years for two trillion dollars. So, in about 2013 or 14 we will reach there we have already gone past the crossed the one and a half trillion mark and from there we are on the same growth path as China. Now, for the economy to grow like this we have to be doing things very differently. So, I tell all the students that stop listening to your parents for instance about guidance in career because their picture is of a slow moving country 30 40 years ago India is totally different how many of us would have predicted the way mobile phones have penetrated the country in the last 5 years even nobody would have. Similarly, computers, software, innovation in technology, embedded systems there is a big need for automation in this in this country now like for instance if you want to if you are a person growing flowers outside Pune flowery culture you cannot grow flowers manually if you want to export you cannot meet those quality requirements for export if you are growing flowers manually you need to have greenhouses with highly regulated environments which can only be done through automation and all this technology at the moment is being bought from outside from the Dutch through Indodutch collaborations through Israel or what have you. And this is not rocket science there is no reason why we cannot do it ourselves even the demography is changing of the country. Buys are very difficult to get nowadays Jadu pocha you know basic cleaning and swabbing in the house we need automation for these things we need inventors to build these kind of things. So, we want to see if we can grow this kind of different in my difference in mindset and our workshop if you like our e-entre project and our course is a medium through which we are trying to effect change. So, this is the slide that I normally start with in my course which each icon represents an application domain for embedded systems. And I invite the audience to consider which of these application domains does not have an embedded system in it. Any volunteers here go on syringe any others I will share with you the kind of feedback one gets syringe handcuffs violin scarecrow is a popular one. Now, let me just talk about the scarecrow. Handcuffs is forensic applications you heard of those students in America who got the RFID tags attached to them to make sure to just track them. So, police and forensic applications lots and lots of applications of embedded systems surveillance cameras automatically checking up on things even nowadays restaurants and malls have got cameras which are doing real time image processing counting the number of heads which are coming in. Face recognition is happening a lot. So, camera is looking at who is coming in and matching the face like you have on picasa if you upload some pictures it tells you which of your friends are there in the picture automatically. Music is the biggest driver of embedded systems scarecrow a friend who is director of an agriculture institute asked me for a scarecrow automatic scarecrow he says I have to imply 4 buys on my hybrid fields and I am paying 12000 rupees a month why cannot we just build a small automated thing which can show away birds. So, my students built it as an embedded systems project a cost project and there is much more one can do with it. There is a company being incubated here called agrocom which deploys little weather stations in fields in the grape growing district of Maharashtra in Nasek and those machines in the field have an effective range of 2 3 kilometers and they are monitoring leaf wetness temperature humidity and helping you predict pest attacks and crop diseases by sending you an SMS. So, like I said everywhere you have this kind of stuff local innovation we been encouraging when we went to Kashmir we said that think of all your local industry needs. How can you improve agriculture or horticulture all these kind of things using technology and do not look westwards for solutions because you might get the solution, but they are too expensive and then you will not be able to compete with those guys. When you just want to absorb technology within the country it is ok to buy it, but if you want to compete with them then you cannot buy it you have to build it yourself and it can be done robots for college education. So, this is the society that we need to prepare ourselves for we want to create engineers who can build these complex machines. We need no problem comes to you with a little label saying that I am an electrical engineering problem or I am a computer science problem it is up to you to crack the problem whichever way you want it can be crack mechanically electronically software wise or what have you. So, just think of the way an iPod was designed or an iPad was designed or some interesting applications like that. It was not designed by this software oriented view that here is a requirement specification now building me an iPad is designed by a cross disciplinary team of engineers who sit down and negotiate. If the guy wants to put a bigger battery for longer use then the chap will say that hey I cannot do that you cannot fit more than this size battery then he has to be very intelligent about the software algorithms he uses in order to fit that much time of use within the machine then you need to negotiate with the display guy and the manufacturing guy right who says if you put that battery will become too fat and you need to make it more thin and so on. So, anyway this is what we need to achieve to create more such engineers and we do it by engaging with colleges conducting workshops and partnering with them and creating and harvesting open source content. So, one of the big problems we had is any college that I had gone to in the past which had interest in robotics if the student wanted to do a project he could not because you do not get a robot in the market and then he had to build a robot to do an application and by the time he built a robot his time is up and again next year the same thing starts again. So, we tackle that problem by designing this robot over the last 5 years is called the fire word robot in this case and we have created an ecosystem where we have a company that manufactures distributes and so on and this is the kind of thing on which you can do a lot of interesting BE projects and more is used as a research vehicle for us also. It has a bunch of proximity sensors in front which can sense up to 1 and half meters to millimeter accuracy it is got a board at the bottom which does all the signal conditioning it is got output in the form of an LCD it is got a daughter board on top which can drive it with a variety of processors you can have a daughter board for 8051 for at mega 2560 for arm 7 and what have you and if all you want to do is give our workshops then you can use these robots which are 2000 rupees one is 15000 and one is 2000. So, you can get started with the robot club using these kind of robots and so on. So, a lot of open source content all the robots are open source a lot of documentation software hardware manuals everything is on the E Yantra project website and when we give these workshops students can go away with a CD with everything on it and microcontroller concepts have become very easy to teach a faculty reflected to me that sir what you have done in this workshop it takes us 6 months to get to it in our curriculum, but in the first 2 days we just hit you with it and it works and we use it to teach a variety of courses like microcontrollers embedded systems mechatronics control systems what have you and this just gives you a flavor of the robot development 5 years ago we saw very expensive robots and then we came up with our own which are the ones on the right hand side and then we ruggedize them and this is the painful process that we have been through a very clue G looking robot in 2005 and we put it into a floppy disk case and then we sort of tried to improve it and finally, we have come to we also tried a lab in a box where you construct your own robot and then deploy, but we found that this does not work with computer science students they are terrified of hardware. So, this is where we reached this is the arms 7 robot this is a variety of adapter boards that sit on the robot with different processes which drive it this is the actuation change to wheels to two tracks instead of wheels and this is with our basic robot board can drive about 4 DC motors or 18 servo motors. So, with that you can have a variety of locomotion you can have the omnibot with 3 wheels and each of the wheels has got smaller wheels on it. So, you can differentially drive those wheels to actuate left right front back base that and the other you can have a hexavod right 6 legged insect kind of thing crab kind of thing. And once we built it we wondered what to do with it. So, we just gave some BE projects and students did some interesting things with it and more important than the robot is the software because software is what helps you solve problems. We have the robot potted under Microsoft robotic studio under you can write C to drive it or you can use Esterel Luster code or Matlab or Scilab right. So, lot of software environments you can use to actually program these things to solve interesting problems. This is what the Microsoft robotic studio environment looks like for visual programming then you can do a physically based simulation of the robot before you actually deploy the code with the press of a button on the real robot. The content factory if you like which creates the open source content is our undergraduate and postgraduate courses where in the last 5 weeks they do a project. So, we have a laptop and a robot for every 2 students to do all the assignments and then they do a 5 week project in teams of 4 right. So, lots of workshops this is just flashing some slides of the various institutions which have been through us right. This is what the workshop with the faculty looks like in the lab. There was an interesting workshop in December 2009 Professor Nagla from NIT Jalandhar came. He went back and gave the same workshop to his students and used the robot that he gave us and they built a nice project he was very excited he wrote he wrote a very nice letter to us and his students built a very cute cleaning robot which they deployed in barber shops to clean up they would have vacuum cleaner and they kind of cleaned the shop right. So, these are the kinds of stories we like to hear from our friends who attended the workshops. So, we have a Scilab plan, but our main plan is how to get students started in robotics how to empower them to solve problems and how to address the challenge of entrepreneurship and all these kind of things. So, I would not take too much of your time now I will just go to the end actually. So, our learning is that we want to spread the study and use of robotics in colleges. We want to build a multidisciplinary manpower and the main metric we assess now to see how much of an impact we made is by seeing the number of BE projects which we have influenced in the engineering colleges right. So, BE projects if you folks are interested in being a part of this Iyantra experiment if you like is not so much an experiment we know it works and it works pretty well now and you want to start a robotics club in your college and you want to inspire interesting BE projects be in touch with us go to the Iyantra website mail me if you like my address is available at IIT Bombay Kavi at cac.iitb.ac.in I am on the computer science website and what I shall do is because we do not have time I shall show you the videos of a few of the projects of the kind of projects the students have done. This is a batch which happened at the end of 2010. So, this is students who have taken the hexapod and they have made it write out some letters that is the first useful thing that we did with it right as you can see h e x and they have learned some very useful things in the process. Then another bunch built locomotion using the hexapod. So, they made the hexapod this climb the obstacle and then perform various dance moves on the top of this obstacle. Yeah in fact it can go more. So, another lot made the robot climb a staircase and once it climbed the staircase it did something interesting right what would students do right they made it dance. You cannot hear the music here we cannot hear it, but they can hear it. So, then it is kind of dancing it looks quite cute as it is right. So, that is what the robot does and another bunch of students built an artist bot right which is it is a robot that we visually captured a picture then they did image processing on it to skeletonize it turn it into vectors and then one robot mounts a marker pen on its pod the camera pod it mounts a marker pen and it reproduces that image. So, without knowing it these guys have built a plotter driver. Now, it is up to other projects to take this code this basic code which is already there and build it into a more complex code like the projects we got after this is some students built a rangoli robot there you mount you mount a rangoli dispenser and it can go to a mall and maybe make a picture by doing a raster scan and dropping rangoli powder right. That is an and there is another group after that which did a project where they used the sensors inside an android phone a Motorola phone in this case to drive the robot. So, by tilt is used the tilt and gyroscopic sensors in the phone to control the robot. So, you tilt it forward it goes forward tilt its side it goes side and so on and they also had to build a bluetooth interface a small board which goes on the robot all these things are done in 5 weeks time right. Then there is another bunch of students who had an electrical engineer amongst them who helped them build a glove controlled robot where you put a circuit board with sensors on it and you control the robot with your hands. The cable that you see there is only a power cable because the batteries do not last very long right. So, so many students are doing their projects here is a tennis ball collector robot where you do image processing to spot a ball an orange ball and then you move the robot towards it right and go and pick up that that ball and drop it in the blue bin ok. So, all this is based on MATLAB, but we have flavors on open source software like Sylab and stuff like this right. So, it is going and it is going to deposit that in the blue box I will stop immediately after this one because I think time is short. So, this just gives you a flavor of the kind of things that students can do and they get very excited and once main thing is to get excited about what you are doing and you do very interesting things right. So, what we are encouraging students to do is to take previous projects and build on it, make them more complex, make them applicable in different domains and stuff like that. There is a lot to do and we have about 100 such projects uploaded on the website, the source code, the documentation, the report, a spoken tutorial which leads you through how to set up and run the project itself. So, you could not want anything more right and we are using this to seed a large number of B projects. See one more thing is students do not realize how important the B E project is right because the B E project is the sign of their maturity if you like as engineers. If a guy has just done say a database application for a library and comes to us right we are less impressed than if he is done something a bit more impressive. So, the first thing that we do when we are interviewing students for projects or for any position here or in for m text and stuff like that is that we talk to them about the B E project because that tells you the mark of a student right what is the level of sophistication. So, what we have done is that we have created an environment with which you can motivate the teaching of computer science and IT and stuff like that using these robots and we have a large amount of open source content there to help you all and the first step is to contact us and then we will take you from there organize workshops and do whatever is required and this is if you like a tester for the embedded systems course which will be running for the faculty through the same medium in December, but that is a separate process. So, this is just to share with you the E Yantra project and to invite you to be a part of it by writing to us. The project website is e-yantra.org and I suggest that if you have any queries or you would like to arrange a workshop suggest you go to the website and let us know. We have a presence on Facebook also and you can check us out e-yantra on Facebook and my email address is kavi at cse.itb.ac.in and do be in touch with us if you want to get started in robotics at your college and we will be happy to help with whatever is required. Thank you.