MEC Webinar the Waterloo RoboHub



good afternoon everyone and welcome to our mechatronics education community webinar today we invited dr. Brandon the heart from the University of Waterloo and he talked about the tour of their new research teaching and training facility the Waterloo rubber house my name is Susanna tajabbar and from concert and together with dr. NEMA lopsy from Southern Illinois University Edwardsville we will be your host and webinar moderators today and now I would like to pass the microphone to dr. Lottery to talk briefly about our mechatronics community thank you very much Donna hello everyone Susannah mentioned my name is Ahmad Lutfi I'm from a mechanical and mechatronics engineering department of Southern Illinois University Edwardsville and I am one of your moderators today so this webinar is one of the many webinars that we have had and we will continue to have in the webinar series on mechatronics education innovation so this is one of the activities that we at the community of mechatronics educators have been undertaking so our community started by it started back in 2016 after a workshop at NYU attendant School of Engineering it was officially launched in March 2017 and since then our first activity was creating a website mechatronics education that org which is comprised of announcements news about the field of mechatronics and robotics a forum and also repository that contains material related to mechatronics and robotics education at the follow-up activity we did we started this webinar series we have had multiple webinar series from well established mechatronics programs around the world and since last year we have also started holding online workshop as I'm sorry in there we are holding workshops in conjunction with technical conferences so we've already held three workshops one in FM EDHEC 2018 robotics summit and Expo 2019 and if EE 2019 which but last week and we are going to have another workshop next in September September 2009 in 19 and Lauren stick so in these workshops we have brought together educators professionals from in the field of mechatronics and robotics education and brainstormed about the future of this field so if you want to be in be in touch with the community and learn about our future activities especially our last workshop and Lauren stick make sure to subscribe to go to this website register to have Annika and create your account there so that you're in the loop of all our activities so as Donna mentioned today we are honored to have dr. Brandon the Hart from University of Waterloo if the title of his presentation is the Waterloo Robo hub what is it what it is and why you should care so let me tell you a little bit background about doctor the heart he is the manager of the Waterloo Robo hub and a sessional instructor at the University of Waterloo he has been working with mobile and legged robots for over 10 years with the research focus on balance and gait for bipedal robot robots in his role as the manager of Robo hub he's responsible for the purchase installation configuration testing of all new robotic platforms and the deployment of any additional essential equipment such as the control center safety systems and environmental variation system he also coordinates and supervises the day to day activities being undertaken by technical staff students and researchers at the facility in general his research span a wide range of topics connected to robotics from 10 the integrity the structures and bio-inspired robot designed to interactive architecture and morphological computation starting early in his academic career brandon has always strived to work on solving problems that will enable the widespread use of robots in entertainment education and exploration we're excited to have dr. Hart as our presenter and I am looking forward to hearing your presentation Brandon hello everybody my name introduced is brandy hurt I am a manager of the Robo hub and I'm also a sessional instructors on campus in mechatronics electrical Computer Engineering and mechanical depending on with a term today I'm going to give you a bit of an introduction to the University of Waterloo engineering program there's a kind of a high level and then talk as well about what the robot is and why you should care and specifically try to link in as much as possible to the education side of how what we do and how we do it improves the education for undergrads graduates and beyond so to start off with Waterloo engineering is Canada's largest engineering school as you can see we have almost 8,000 undergrad students and over 2,000 grad students and over 300 faculty members an almost an equal amount of staff supporting that what that means is that we can do things that other schools in Canada or I can't really do because we have that critical mass and one of the things that we do well is kind of this innovation umbrella that everybody kind of talks about and the main three reasons that we can actually achieve that is because at the University of Waterloo we have an inventor owned IP policy which means that whether you are a staff member a student a professor or postdoc whatever it is if you come up with something it is yours if you'd like to loop in the University and have them take care of patenting or generating you know helping you generate funds things like that then you can give away some of that IP if you'd like to but there is no requirement for you to do so so anything you come up with invent create etc is yours and because of that we have a very strong startup culture which I'll talk about on the next slide where professors are part of startups students who are coming out of undergrad coming out of grad studies immediately startup companies and can create things and use the developments and unique pieces that they've developed to start companies in the area and and abroad really the other two pieces of that are the transformational research which the robohelp is a big part of and we have a huge research program which is just growing exponentially with a very heavy focus on engineering we we have significant investments both from the public sector in the private sector in effectively all all fields of engineering across the board because the University of Waterloo was started as an engineering school that right from the get-go wanted to be connected to industry which ties into the third point of having a heavy emphasis on experiential education so everybody who comes in to do an engineering program must do at least five co-op terms six or more if they choose to and so potentially you when you finish your engineering undergrad here at Waterloo you could have worked for six different bosses at six different companies doing six different things and you have a reasonably good idea of what you enjoy what you don't enjoy what you're good at you're not so good at and the kinds of technical expertise that you might need for the types of jobs that you want and it also prepares our graduates very nicely for the outside world and I'll just give you a couple of details about that before I move into the actual meat of the presentation about the lot of the robot so just a few facts here on the entrepreneurship side and the startup side we have hundreds of entrepreneurs that have raised billions and billions of dollars as of a couple years ago and that number has obviously gone up since then and we have velocity on campus which is actually the largest free student incubator in the world with a relatively high survival rate especially for start-up started by undergraduate students for the most part and we are also in kind of top lists for being vc-backed creating unicorns creating jobs all those kinds of things so we are very much involved in our community involved in that side which then leads to a lot of people coming back from those startup companies or even from larger companies to talk to us about both entrepreneurship and about the kinds of research and education that we provide so we actually have something that we're working with a little bit called reverse co-op where people in companies executives managers etc can come back to the University and learn about AI and how it can affect their company or learn about robotics and how they could potentially put a collaborative robot on the desk either their desk or the desks of their employees and how that might Ottoman the capabilities of those those employees and then the last kind of general slide that co-op is extremely valuable it is a huge part of what we do here at Waterloo particularly in engineering but actually across the board most of our programs at Waterloo have a co-op component of some kind many of them are optional as opposed to mandatory like it isn't engineering now but it gives everybody a kind of that good basis so they can take that and move into the world with some real real experience at companies like I said with multiple bosses things like that and a lot of students coming in actually choose Waterloo because of our co-op program because we are like it says number one in Canada and number four in the world for employer partnerships and because of that we also have a lot of Industry connections coming in who then work with our researchers work with our professors and instructors lecturers to incorporate the real-world experience into the educational system and into the research program so it's very much kind of an interactive thing which you'll see as I kind of go through the slides about the Robo hub specifically okay so the Robo hub Waterloo Robo hub is a state-of-the-art or products research and training facility with a globally unique fleet of robots what does that actually mean it means that we have built a relatively large room with many many robots in it lots of different types of robots which you'll see in a little bit which is dedicated to research and training so it's not owned by a particular professor it's not beholden to a particular fund or degree program or department or anything like that it lives under the Faculty of engineering so we're kind of a general space and we have in any given week anywhere from 6 to 40 or 50 professors and they're postdocs grad students undergrads undergraduate research assistants fourth-year projects all kinds of things working in the robot specifically the Robo hub is basically working at the intersection of collaboration and robotics so in any particular combination of those we are interested in looking at both the research themes the connections to industry how it affects education and how our students and our researchers need to be connected into that ecosystem in order to understand how things are going to work so that might be how to make robots work with robots how to make robots work with humans how to get academics and industry working together on a project one of the hardest ones is getting professors to work together and I can say that because I'm not one really it's getting all of those bits and pieces together so that you can look at every level of collaboration when it comes to robotics and we also work very closely with the AI Institute on campus we work with the autonomous vehicles team some of the student teams lots of different sides of the coin because we have kind of this central location in a new building and we also have a mandate effectively to look at all of those connections between collaboration and robotics what does that look like on the research side effectively we have these four key research themes so obviously we have a lot of robots that are at the early stage of research so we have to look a lot at the Oh side both high-level and low-level control so for something like our industrial level full-size human eye that you can see in the back of the Talos control is a big part of what we're doing and what we're working on for a system like that moving into some of the more common ones like mobile manipulator or drone things like that we look at some of the planning some of the perception specifically looking at multi robot planning and shared perception so sharing maps between systems having one robot which might not have a good view or a good set of sensors for particular thing able to request information from another one so maybe a ground system doesn't have a good idea of what's in the next room so a drone will fly in and give it almost like an eyes in the sky and kind of approach and share that map share that information so that the ground robot doesn't necessarily have to have a full depth camera and all those things you can ask for that information from another one of the robots and then obviously to tie everything together and tie the humans into the loop is that interaction piece and that is robot robot interaction robot human interaction both on the interface side on the control side how do you actually program and how do you teach it to do things if you want to run some kind of AI in the backend for vision for a pattern recognition of commotion learning all those kinds of things so those four pieces are kind of our core pillars that hold everything else up and in order to look at these types of questions we actually have acquired a globally unique fleet of over 30 robots so as you can kind of iconically see here we have humanoids arms and legs we have fixed base arms we have flying drones of various sizes and scales we have autonomous vehicles both indoor and outdoor and various combinations thereof so mobile manipulators with two arms mobile manipulators with one arm some of our drones have multiple sensor suites on them and coordinate with ground platforms to charge things like that so a very interesting set of tools that we can use to look at all of these various bits and pieces so I'm going to quickly go through a few of those first off we have obviously the smaller kind of human interaction social interaction robots on the human right side but we also have industrial-strength humanoids like the Talos back here and we're looking at a lot of different things again to augment humans so carrying something with the human and in this case there's no actual perception going on so all of the responses with the table are based on torque sensors and the elbows torque sensors and the arms it wasn't actually watching Alex who's in this video or a humanoid specialist what he was doing it was only paying attention to how the table was moving and the compliance they can see there is enabled by the torque sensors that are in the arm in the upper body so it can fully balance using a lower body controller which is distinct from an upper body kind of softer controller where it's only trying to spring its hands back to a particular location relative to its body as opposed to rigidly holding one firm which gives you a bit of that collaborative safety and it gives you a bit more of a sense of it effectively gives the robot a sense of feel so it doesn't have touch necessarily like it's not coated in the skin or anything although that can be done but it does have an idea of how hard it's pushing on the world on people and how hard they are pushing on it and one of the reasons we use the small humanoid robots is to map some of the social aspects and the human interaction aspects on the small robots where we can test it and it doesn't really matter if something breaks or it swings its arms around it's not really going to hurt anybody it can't hurt anybody we can map those human interaction lessons that we learn up on to the bigger systems like this one the next piece is the collaborative robot arms here we can see a couple of them Frank Amica panda on the right and the Sawyer on the left which has been bought from rethink by Han but really the the similarities here and we have several other fixed-base collaborative arms as well the big thing is that they have those similar torque sensors or in the case of the soy or cereals elastic actuators so that they know again how hard people are pushing on them how hard they're pushing on the world you can see there's actually a hand on the panda on the right there guiding it so you can actually teach it by dragging it through a set of poses and using buttons on the robot itself to teach it much like the robots they're the buttons that are on the Sawyer on the left as well so you don't actually need to be a robotics expert to teach these robots whatever task you want them to do you can bolt it to the table you can teach it a task it can do that a hundred times and then you teach it something else and you can adjust that using a super simple tablet interface right out of the box and then you can also for all of these systems go deeper and deeper as needed to satisfy the educational and research needs of the students and professors and lecturers and everybody who is involved with these systems so you can have that initial human interface which is relatively simple for someone where you want to treat this as a tool that gets planted on your desk and you're using it to do the dull dirty dangerous boring repetitive pieces of a particular task so you're effectively treating it as just another tool in the toolbox and then if you get further into the education side you can look at okay well how are they doing how do we modify that these are redundant robots so they have more degrees of freedom than they need more joints than they need so how do you figure out what to do with that extra motion freedom all of these are kind of questions that we're looking at in our undergrad and graduate education programs particularly in mechatronics and we actually have a fourth year design project going right now that we're supervising where they're trying to figure out a way to build one of these effectively collaborative robot arms with torque sensors and seven degrees of freedom to get the redundancy at a scale that you can use for education where every student in the class has a robot as opposed to something like these where you might have two or three for a class if you're working combining that a little bit you can get collaborative arms that actually mount them to a mold manipulator like these two systems where you can have different types of systems different types of arms two arms one arm sensing potentially there's a camera on board or in this case there's a remote camera telling it where how those things are being connected together and these will enable you to take those collaborative arms and some of the lessons learned there with having it bolted to the table and then apply that to okay well now you have a vehicle that is capable of moving throughout a space so what challenges does that impose what benefits so you do get out of that and there's many different aspects to this there's the sensory part there's the perception part of where is the table where is the robot how is the arm moving there's also the controls piece is trying to unify the controls for the mobile base compared to the arm and figuring out obviously this isn't interacting with humans but if you have an arm on a mobile base like the mobile the black one that we saw first how does that interaction actually work what does the interface look like is it voice is it a tablet do you have you know a keyboard somewhere and that is how you're actually controlling and interacting with it because all of those things will change what kind of sensory modalities you need what kind of background you need to actually program a system like that and how many layers and what kinds of layers of safety and control that you need to make those kinds of things work so moving on from kind of the mobile manipulator side of things we also have a relatively diverse set of aerial platforms so we have an outdoor system that you saw me adjusting the camera on there and we have several of those and we also have several of these cones of cue drones which allow us to do a lot of nice indoor research multiple platforms we can combine them with ground vehicles and get kind of right out of the box they work well with they're kind of paired ground systems so we can have them follow each other as it's doing here we can have one drive the other we can have all of them be autonomous and we can look at those kinds of systems we can have multiple ground platforms multiple aerial platforms and generate algorithms and equations and planners and all those kinds of things for these systems that are indoors and then map those to maybe bigger outdoor systems for example and you could look at things like when you're in a GPS denied environment with outdoor drone how does that work maybe you have a base station on a big mobile system like we'll see in the next slide which can carry your power it can carry a GPS base station or ultra wideband localization system something like that so then the drone only needs to know where it is relative to a ground platform on the ground platform if it loses GPS it knows that it hasn't left the ground for example it's not gonna fall out of the sky when it loses GPS it'll just stay where it is or try to find GPS again so you can use the strengths and benefits of the different platforms together because we have all of these various types of systems at the robot to combine and extend the capabilities that you would have with only one or two of these systems and we can map much like we can map things from the small now humanoids up to the big Talos we can map from these huge runs and Cuba it's relatively simple to use very reliable kind of education systems Cuba and cue drone to bigger systems outdoor systems more systems all of those kinds of things and the one of the nice things about these on the educational side at least is that you can run MATLAB right on it so if the students are already learning MATLAB in the class which they do an undergrad here they can directly effectively draw the control system that they want and simulating for example and they can actually run that right on the ground platform or the aerial platform have them talk to each other all that good stuff you'll notice also that these have markers on them and we will talk in a bit about our indoor positioning system but one of the nice things about that is that we can use these to validate algorithms that are running either aerial or ground systems and for planning for perception all those kinds of things and we can make sure that the maps and distances and all those things that are being calculated we can verify them with the indoor positioning system or we can just use the indoor position to the system directly and we don't have to build in a lot of those additional pieces so it makes it really nice on the education side that you can mix and match which components any given student or group of students is actually looking at has to develop themselves and one of the things that we are doing in the robot app is trying to make it so that when a new master-student let's say our new undergrad student comes in as a co-op or something they can ramp up really quickly on the platform that they need for their particular problem and get right to trying to solve their problem as opposed to spending a term finding a robot in another term finding funding and another term learning how it works and another term getting it reliably working and flying or driving whatever it might be all of those things we are effectively taking care of with our staff and our knowledge and understanding of all the various systems so that when they come in they can say okay here's a few weeks of instruction and tutorials now you're ready to work on your planning problem for example and one of the reasons that that's super beneficial is when we go outside so these are a couple of our outdoor platforms our warthog is actually orange but it is still in Phidias it's got a bunch of sensors on top multiple GPS setups and a valid ein laser scanner those kinds of things and we actually use these systems to both upscale some of the algorithms that we run inside and to do a little bit more of the general kind of dirty or dangerous work that we can look at outside we can simulate those kinds of things or actually do them outside because these are fairly rugged systems and the vast majority of the systems everybody talked about this yeah but the vast majority assistance that we have all run on the robot operating system on Ross so because they have that common middleware generally they can all talk to each other you can run the same planner on multiple scales of system things like that so we're trying to make it really easy for people coming in to understand how to take something that they've learned in class and map it onto a real robot or how somehow they take something that they've built for a husky like the one on the on the left and map it up to a bigger system like the Warthog on the right what changes what stays the same and how those things how and when those things need to be paid attention to effectively I'll move a little bit move to a slightly different topic here so we also one of the unique things about the robot is that we have professor there at Kemah see here on campus and he is a world expert on magnetically systems specifically not like a maglev train where it's very much linear you have one track and it travels along that track but actually in 3d localization 3d motion 3d levitation of initially micro magnetic robots and now this platform here is a little bit larger I'll give you a few details on that so this is basically a magnetic table which is approximately 2 by 3 feet and there's actually several different types of modules I guess that we have for this one the first one you saw was just purely capable of lifting up and down it only had one permanent magnet this one you see on the right has three different permanent magnets in it so we can actually control the position and the orientation roll pitch yaw fully with this system and the way that we do this is actually by using a set of individually controlled electromagnets in this table to modify the magnetic field above the table in order to manipulate how and where these magnets are and you'll see again those reflective balls and in the picture in the top some of the cameras that are used to actually localize those those reflective balls there I'm not going to go through all the details here but basically the applications for this size of system are for cleaner manipulation things like that because there's no moving parts where the actual module is so within this levitation space within the volume there's no moving parts there's no need for grease or there's never gonna be rest there's never gonna be any of those things unless you're introducing those and another and another way so effectively you can create a confined cleanroom and use this to do six degree of freedom effectively full 3d manipulation of an object of multiple objects as needed you can also put it in a wind tunnel obviously because there's no support system for this you could actually look at what the forces are because you can measure that based on the magnetics pick-and-place collaborative operation etc so there's lots of different bits and pieces to this and you could because it is again just magnets in here have liquid and have the motion of the actual module itself help you to measure so you can actually use this magnetic module this maglev module as a sensor so you can look at weights you can look at density and viscosity and all those kinds of things and then if we move from this up to the full-scale one that we are currently designing and installing so in some of the previous pictures you might have seen the tiled floor in part of the row Bob which is outlined in that top picture and basically in there there's going to be a 2 by 3 meter levitation volume which is not super deep but deep enough that you can do a lot of interesting things with it and we can also have multiple platforms running at the same time so because we have individually controlled electromagnets underneath the floor we can actually have them move around independently we can have them do coordinated tasks and here obviously they're being treated as simple floating systems but you could do any number of different things with this again reasonable load capacity 5 kilograms and moving at a relatively good speed as well and this kind of larger system which will be the only of the only one of its kind because if it's being developed and created here on campus specifically for our our space and it enables like shown here some flexible manufacturing you can use it to carry things in a warehouse if the magnetic volume gets longer bigger etc and again you can do some more of that cleanroom manipulation confined space manipulation and many of the things that the smaller table is also capable of obviously you can do with the bigger table it's a little bit more challenging because with the bigger table your resolution on the magnetic field gets a little bit lower obviously but the accuracy and the positioning and all those things are relatively the same and if you want to scale up it's basically just a matter of cost in terms of how many amplifiers you won't have any electromagnets etc because really they're they're just analog systems that you can define how strong you want the field in any particular magnet electromagnet and then have these platforms that you can localize using whatever system you want whether it's video or an indoor positioning system or anything else so getting a little bit to the outside level and in the Robo hub we also have a few things that enables to do some interesting work on humanoid gait and measurement validation of sensors testing of external robots all those kinds of things and a big big main one is that we have an advanced vikon IPS which is can also be used as a motion capture system we currently have 24 cameras we're hoping to approximately double that maybe not fully double because we also want to put a set on the ceiling so we can use our full volume for collaborative flight and swarm flight research in tandem with this we have a control center that we use for real-time fleet management and visualization so a lot of visual feedback a lot of details on what kind of messages are passing what the latency is all those kinds of things you can analyze a lot of a network side of things because of a VNA so you can look specifically at the wireless information and density bandwidth all that fun stuff the window is actually on the Robo hub our privacy glass so at the flick of a switch we can turn off all the windows so that instead of a showcase space where you can see everything that's going on you can watch research happening in real time watch the robots moving around you can turn it off if there's a confidential project or if you want to create an artificial environment because once the windows are nominally quote-unquote turned off you can use those windows then as a projection surface you can project an artificial environment you could project sensor data information for the people information for the robots etc and we're actually looking at getting some augmented reality and virtual reality components as well in the space so that the people that are in the space can actually see what for example the shared map might look like so if multiple robots are collaborating they're building a map together you can pop on something like hololens potentially and look through that to see what the actual robots map looks like and understand why they're running into an obstacle maybe they can't see it or they think an obstacle is a different size because of reflectivity or you know maybe there's a sensor problem potentially you can see that right away and get that intuitive feeling of oh now I know what the problem is I can fix it you're gonna have to dig through you know a 3d rendering on a computer and try to figure out how those connections Burger sorry and one of the other things we do is we have various fans stairs doors rubble etc so we can do real-world tasks simulation in the Robo hub and make sure that everything is working how we expect that the algorithms are working as we expect and the physical robots are doing what they're supposed to before we then take it outside into a windy environment or try to make it do stairs or open a door et cetera so we can actually do some of that railroad simulation inside which is especially helpful because we are in Canada so lots of time there's snow or rain outside and so it's nice to have a space that we can simulate some of those things that we know is somewhat controlled and we can verify some of our assumptions before we bring systems like the word hog with us key or some of our drones outside or out into the world to see and as the the real-world impact and one of the big emphasis pieces we have is trying to look at real world problems and find the solutions and then pass that along through courses through workshops etcetera to the students that are coming in the grad students etc and the last piece that I'm just going to mention is that because the vast majority of our robots are based on Ross the robot operating system that I mentioned we also have an initial version of a multi master network in development and we're debating right now how we should do that for robots that are not actually running Ross so whether we use something like quark as a back-end to hook all of them together and then that produces Ross notes for those various things like the maglev system that we're defining or if we go to a model like Ross – where you can have multiple systems and you don't necessarily need a master they can talk to each other and so that so we're actively working on that right now and if any of you are actively working on that we would be very interested in working with you on that aspect with that I think I will close up and say have any questions please send your questions through the webinar questions page Thanks Thank You Brandon for a really interesting presentation we have received a couple of questions from the audience and as Brandon mentioned if you have more please feel free to send send them through the questions window we have some 15 good 15 minutes to answer the question so the first one that we received was about the products or quite a few are about the products that you have in the in the Waterloo rubble hub do you have a universal robot product no we do not so on the collaborative robot arms side we have two Sawyer's from rethink which are now on we have two pandas from Frank Amica we have two Jacko's and a gen 3 from Canova in Montreal we have cuca by WR 14 which is the bigger collaborative robot arm from Koopa and I think that's in terms of arms because one of the things that we look at for when we're acquiring system specifically for when we're requiring arm systems is that we'd like to have and the capability to actually measure torques and forces in every single joint which all of those systems allow us to do the you are arms do current based torque sensing which isn't reliable enough for the the kinds of research that we want to do with collaborative robot arms here and the most of the robots and equipment that you have is of the shelf or they also build robot manipulators or rather so as of yet other than a few fourth-year projects were supporting um which is kind of a capstone project for the 4th year students here we haven't built any custom robots we have customized some of the robots that we bought off the shelf and so some of our clear path robots for example we put you know different sensors on them or mounted an arm or just mounted arm various things like that we will be as we move forward working with Professor dot and hon Kristin Hahn who effectively created the field of social robotics to see if we might want to develop a new version a new iteration of the social robot that she worked with at Hertz in the UK and I have a feeling that some of the professors on campus who work on exoskeletons are going to be leaning on us a bit to potentially develop or modify exoskeletons for them but part of that part of the idea behind the Robo hub was to be able to figure out if you don't have someone who can build new robots develop new systems etc if you're buying things off the shelf how do you make them work together how do you make them work with humans etc because if you are a company or you're a smaller educational institution you're going to want to just buy a robot or several robots which ever apply to your particular task or kind of goals and have them play nice together which is part of what we're trying to develop and create thank you do you use any robot simulators such as VI Borg Zeebo to test behavior of the robot manipulator so because we're we're working mainly in Ross we do use gives Nebo a fair amount and we are working right now to try to figure out if there is a server-based solution we can used whether it's AWS troublemaker or the construct or what it might be and all of those use gazebo or some form of gazebo which for most of our robots because they are Ross native that's kind of their their natural habitat we have been also looking at Z rap and I can't remember the other one but most of the time we're playing in gazebo and because we are working with raw space robots thank you so a couple of questions now related to the Robo hub operation and structure how long has the Robo have been in operation and do you have any statistics already that can line the outcomes with respect to quality of undergrad education so the Robo hub itself or the whole building that should go the entirety of engineering 7 which is where the Robo hub is located we only go on acts in September so it's been open for about nine months now ten months I guess and the official opening was at the end of October so nominally it's been kind of six to eight months of actual useful operation throughout that time we've been acquiring robots pretty steadily and so I think the last robot that showed up was only about a month ago and we are finally getting to the point where other than the maglev system we're getting close to having all the robots we want and now we're working on ramping up our end of environmental simulation and additional sensors additional cameras all that kind of kind of stuff and so because we're so new we don't really have any statistics that we can point out or look to in terms of how impact is affecting or education is affecting our impact or vice-versa and but we do have requests it seems on a weekly basis from undergrads and graduate students both from our campus and from campuses across Canada to work with our robots to come in and kind of get a workshop with us things like one of the things that I didn't really mention when I was giving the talk is that the idea behind the robot once we're kind of up and running which should happen fall at some point kind of fall fully operational if we want this to be a resource for all of Canada because the lab was started it was kind of founded with a CFI grant was the biggest investment in robotics at the federal and provincial governments have made in over a decade and we want to be able to shader and of that wealth of infrastructure the wealth of equipment and experience with everybody across Canada because we know there's a lot of places where you don't have the space to put up a service a lab like this or maybe don't have the funds to get something like Talos or some of the bigger systems you might have one or two so one of the things we're working on doing to kind of tie a few of these questions together is to actually make a how the cloud-based system where you can simulate and test behaviors and things like that online and then you can actually send that up load it bring a person to the Robo hub and test it on the real hardware that we have here we want to make it very much kind of a physical and digital resource for everybody to to do robotics education and robotics research thank you you touched firstly on the next question about the local hub funding but on top of the governmental fund you also partner with sponsors particularly in kind sponsors yeah so we do have there's there's kind of three pieces to the the model for the Robo hub so it was started by a CFI grant like I said which is partially federal and partially provincial and then partially in-kind so part of the in-kind came from the University from the Provost office and part of the in-kind came from companies like clear path and Kwanza and pal and those kinds of places where when you use CFI funding to acquire robots there is an expectation that the company will provide some portion of the cost as an in-kind donation effectively we also have funding from the advanced manufacturing consortium which is an Ontario group of three universities which the university is part of it and the Robo hub is one of the four labs on campus that are part of that group and they were the ones who funded most of our collaborative robot arms in some of our smaller systems that we use to look at kind of industry problems manufacturing problems and those those those kinds of things and we are moving forward looking at developing a model for partnerships with institutions and with companies where we can do other in-kind things we've had a few companies who have literally just sent us robots and said here please use this as well as partnerships for training for doing that kind of reverse co-op piece that I've been talking about we're trying to figure out what that relationship looks like but we should have that figured out by hopefully the end of the summer and then we can start offering effectively membership in the Robo hub in exchange for training or or kind of using our expertise maybe having some time in the Robo hub to do experiments that said are those tens of great how many staff members equivalent full-time members are tied directly to the Robo madman currently we have four and a bit a little over four full-time equivalent members our staff because we have myself and then Alex the Human Rights specialist that I talked about we have a full-time technician that just started this month and we have a postdoc who just started this month he's dedicated to the robot hub itself but we also have a support team from our pis and the other professors they use the space because we have up to 40 or 45 faculty members and they're postdocs and their technicians their lab managers working with us as well but dedicated to the robot there's four of us are y'all for doing that curriculum development for the college students for new students onboarding yeah so that's one of the things we're working on throughout the summer actually is figuring out both the safety piece and the training piece so for doing the onboarding figuring out what training they need to work with particular systems raus workshops which were actually we have partnered partnering with clear path to and kind of come and run several kind of beginner Ross workshops as well on campus and we're gonna start running a almost like a Lunch and Learn session every couple weeks and the fall we're hoping to start that and where we're gonna look at the practical aspects of doing robotics and those will be open to everybody undergrad staff graduates how do you manage usage of Brava have in terms of scheduling what are the hours of operations is good and have students have access after hours so currently because we are still working through all of the the safety training aspects somebody with access to the room which is restricted to employees of the university so faculty and staff basically in postdocs and has to effectively let you into the lab but if you schedule a time we've had experiments happening anywhere from Adm down to 8 p.m. really and some on the weekends we've had events come in on weekends and things so it's really it's really dependent on on the needs because we also host a lot of tours for main campus for the dean's office for the department etc and whether it's government officials coming in or industry or visiting professors etc and a lot of those things are actually also critically important because part of what we're doing is trying to make those connections and so we kind of cooked in in our mandate that we want to be able to enable those connections and enable those kinds of tours and things to happen as well and so there's a bit of hard and fast rule about access it's really more about figuring out how much time you need and what time you actually need in the Robo hub because we have a lot of simulation tools we have a lot of remote access tools and things so most of the time if you're not physically working with a robot you don't necessarily have to be in the Robo Hobbit cells and this also ties a little bit to be the next question about the external access one question is what are some of the ways to connect or meet up with mechatronics students interested in co-founding robotics startups and also if you have already be any process for access to robohelp for external students or you are still working on that so in terms of general access we don't have a process yet if you go through one of the P I is one of the professors who's involved directly then it's fairly easy to get access because effectively they're kind of sponsoring you to use the space we do have a lot of undergraduate research assistants from first through fourth year we have co-op starting soon I think we have a one or two in the fall and we have visiting students from actually coming in the fall as well who have been sponsored by one of the PI's to basically come and do their final year capstone projects in the row pub hosted by the university so really right now it's a it's a question of figuring out what kind of relationship that you want to have with the Robo hub and with our professors and then just kind of finding a way to make that mean you can work in the Robo hub with the robot cetera and we are planning on making slightly more formal structure and at least on the industry side less so on the academic side so that we can say you know as part of your membership you get and for people to reverse co-op every year and you get you know three days in the robot or two days in the robot whatever it is and so we're working on that and we should have it and by the fall in the fall at some point but but yeah as of yet we don't have a kind of a defined process for doing that one question back to your your tools in robohelp have you ever taken the q drone outdoors I hope you say what I hope you say but I would like to spin it a little bit which of the robots you actually can use outdoors and what year if you are flying any drones outside what are your challenges there's given all the permits or you know do you have any expansion of the robot out outside yeah so we have most of our clear path robots are indoor or outdoor so the Warthog the Husky is the Jackals and the ridgeback is on holonomic wheels so it doesn't go outside but those are all on the ground so there's no real rules about that other than hey can I Drive in your parking lot or a gravel pit your farm your etc and when it comes to the drones the cue drones we have two spaces on campus both indoors that we fly them so we have the Robo hub and we have a flight arena and one of the the older engine buildings and and then the outdoor drones that we have we have a couple met Reese's and we're getting an Ariane I Ranger I think it is that we do use outside and those need a lot of kind of paperwork to be generated to actually do that reliably and safely particularly because of the new Transport Canada rules you have to kind of register the drill and you have to do you know if you you have to get registered as a as a pilot with them and there's there's a lot of different things that we're going through and we're trying to figure out a way to streamline the process right now so that anyone can run their code on the drone but one of our registered license etcetera pilots for lack of a better term will just have to kind of accompany you and the drone and be in charge of the kind of remote takeover and all that kind of stuff so that the students themselves whether they're visiting or otherwise don't have to go through the whole process they just need to be accompanied by someone who has so that we don't have to jump through all those hoops for every single person who's gonna want to put code or a sensor or hardware or whatever on on one of our outdoor systems I would like to thank you for really great and interesting presentation and thanks to everyone who joined up today I would like to again mention the upcoming workshop of the mechatronics education community that will happen September 27 to 28 at Lawrence Technological University so if you are interested visit the mechatronics education website for the details that will there are some already but more information will be coming in the next couple of weeks thank you very much for joining us today

Be First to Comment

Leave a Reply

Your email address will not be published. Required fields are marked *