Startup Battlefield Finals – Scaled Robotics

Good afternoon everyone Construction impacts our daily lives in Unique ways it shapes the cities we live In the homes we dwell in build the Infrastructure that drives our economies Yet this Global 17 trillion dollar Industry also one of the world's most Polluting is one of the world's most Wasteful and inefficient And we believe this is because whilst we Design digitally we still construct Manually construction is inundated Without data tools and processes they Just don't have the technology to keep Track and measure what they're trying to Build They use chalk bits of string to Construct and Post-it notes to keep Track of progress on multi-million Euro Projects For example On a site in Oslo last year a structural Beam was brought to the attention of Site management This was three centimeters deviated from Where it was supposed to be built and Entirely missed by existing quality Control including the site survey If this had been left undiscovered would Have cost thousands of Euros seven days In delays and the valuable time of eight Site managers and that's just one beam And the crazy bidders this is considered Normal in construction and this is the

Problem scale robotics is trying to Address with robotics and artificial Intelligence we're building tools that Can track and accurately measure Construction progress so that your Projects can stay on budget and on Schedule It's simple you can't manage what you Can't measure And that demon Oslo that I told you About Automatically flagged up by software Developed by scale robotics under 24 Hours brought to the attention of site Management and fixed with no extra cost So how do we do all this Our robots drive around site building Detailed 3D maps of the environment this Catches the current state of the Construction project We then automatically upload this data To our servers for analysis Switch to demo please Then we perform non-rigid registration Non-linear optimization machine learning On the raw data to filter out all the Noise and provide precise statistical Analysis regarding the quality of build On every element And this complex information is then Delivered back to the client on an easy To use web-based interface What you see in front of you is simple Color-coded information gray means we

Don't have enough data to make an Assessment red means it's missing green Means it's been built correctly and Orange as you might have guessed means It's built incorrectly And as you can see we are standing in a Sea of orange which means everything is Wrong Now knowing everything is wrong is not Particularly useful information as a Site manager would you care most as a Site manager is what's putting my Project most at risk and we make this Information searchable Let's say you've had issues with quality Of Steel work you can say show me all my Beams and columns Filter them By date of build Not only that filter them by quality of Build show me everything that's deviated Above X millimeters Now what you get is a automatically Ranked list by deviation so you can Focus your attention on what is putting Your project most at risk rather than Having to sift through hundreds of Thousands of data points in fact you can Even go further and diagnose one of These issues You can visualize Pointwise deviations with a heat map And now we've gone from whole scale Site-wide information to extremely

Granular pointwise information But if you're like me you want to see This with your own eyes so we've Automatically tagged every element on Site in every image that it appears so You get multiple Vantage points to see What's going on But like you can see all beams look the Same you can go ahead and highlight this Element And now you know exactly what's going on With your site And the really cool bit is this complex Analysis that we've just shown you is Available to all the stakeholders at Their fingertips even before setting Foot on site Switch back to presentation please So we are taking this to Market on a B2B SAS model We integrate seamlessly with existing Workflows and existing laser scanners We also integrate with all widely used Industry standard file formats Our typical client is a large general Contractor with multi-billion Euro Revenue we onboard them on a single Project and then grow within the Organization We're running active projects in Europe And I've done tests in both Middle East And in Asia we also want to thank our Partners at Autodesk who have the Largest market share in AC software they

Have championed our product to their Customer base Every product needs a great team we're An Eclectic mix of academic and Industry Experts where more than half of our team Has phds our CEOs to it is a trained Architect with more than a decade of Experience in the industry construction Industry and I have a PhD in robotics And artificial intelligence and I spent My past decade at some of the world's Most renowned labs in both industry and Academia So to all the large General Contractors Out there are you tired of one and two Percent margins Then shoot us an email we're confident Our product will help you stay on budget And on schedule on your next project Thanks What's uh coming out here yeah Cool thank you for the pitch Um So construction is a very uh a Notoriously large and non-digitized Industry and I guess the nature of many Projects is that they're very custom and Quite different Um so how are you think about Introducing automation both in terms of Data acquisition and interpretation Rather than doing um a custom service Every time you have to analyze a new Project

I mean every project is unique in its Own way but there's so many things that Are not unique we still use concrete we Still use steel predominantly most Countries have kind of a set way to Construct most buildings so yes every Project is put together differently but You're still pretty much using the same Materials on every single project so What we learned from one project does Directly impact then the the information That we can deliver on the next one so There is quite a lot of uniform details There we take it to the basics at the End of the day you have a 3D model You're trying to manufacture it on site It is on-site manufacturing it's like Any other manufacturing industry and What we're doing is we're taking lessons Learned in other manufacturing Industries where they've gone from a lot Of ways to zero waste like cell phone Manufacturing and that's because of Process control and lean manufacturing We're trying to bring the same thing to Construction And how does a relationship work exactly With the customer you said that you're Charging them for a project do you lease Them the the hardware how long does it Take a machine to to go through a site Do you need a human that you provide to Then maintain the machine or bring the Machine or take the machine how does

That Dynamic work so we do it on a Monthly a monthly fee so depending on The size of the project we can put one Or multiple robots so it's a scalable Solution hence the namescale robotics so If you're if you're doing something Maybe a hundred thousand square meters Maybe you're going to require two or Three of these robots and you'll Strategically deploy them around the Project we then charge for a monthly fee On the amount of data that you're Producing the amount of robots that you Have Um and what kind of analysis that you Want to get out of that and then in Terms of the people we reduce the amount Of human interaction with the robot but There is a certain relationship that the General contractors have to be able to Direct where they actually want to Capture that information on a particular Day so there's no point doing it blind And letting the robot drive around and Capture something that's not in the Schedule so it's kind of a hybrid Relationship that they have between the Two okay gotcha Um I'm having a little bit of a problem Here in the sense that what's the real Advantage of having a robotic solution Versus say allowing Um site managers Architects to use Commodity cameras to capture the data

Upload it to your platform you don't Have to get into Hardware you don't have To worry about you know driving a Machine around or whatever the scale is The fact you have a platform which they Can use and then you can do you know Visual uh you know interpret the data The visual data yourself clearly which Clearly you were doing already so why do You need the hardware so there are two Parts to that answer the first thing is Our software platform is agnostic to the Data capture device we can take any Traditional laser scanner and that data Can be uploaded to our software platform And we can still process it in fact a Couple of the pilots we're running They're using their own laser scanners This is where we seamlessly integrate Into their workflow they don't have to Change anything about the way they work To start using this as for all cameras If you just do rock cameras you are not Going to get the same Precision that we Get you do need a laser device to be Able to capture the data with the kind Of precision and accuracy that we can Provide And you need to you need to produce your Own uh laser device and cameras to in Order to get get the accurate data no we Can actually still work you can use our Platform too ours is a faster reality Capture platform but you can also use

Your own laser scanner we are providing You the option of using our platform but You can use your own laser scanner too I See okay great totally understand that The construction industry has a ton of Waste and a lot of that waste comes from Error and think catching Things Early Seems like a very significant value Proposition how do you quantify that for Your customers and do you have case Studies or examples where you've been Able to pinpoint a specific waste Reduction so we're developing those case Studies now as you can imagine it takes A little bit of time to kind of see Through a project that maybe lasts 18 Months to see every single detail but Give you a a clear case example the First day we deployed on a construction Site I will not name the company or the Location the first day we deployed the First time we took data we found Approximately 10 to 20 000 Euros worth Of mistakes or things that would of Course issue and later on down the line They got fixed within that week at no Extra cost but the longer that goes on The more complex the solution to solve It is so there was a structural wall Which was 10 centimeters from where it Should have been if it was caught on the Day they can just knock it down and move It without any problems because there's Nothing on top so there's no structural

Issues one month down the line that Schedule delay is now going to impact All the subcontractors that rely on that Space to be able to work so the longer It takes the more it's going to cost Yeah and just a follow-on so I mean in Some respects you show people you reveal We're PR where poor work has been done Or where people haven't followed process And so people might dislike this or they Might put their angst against this uh Beautiful piece of equipment how do you Engender love and instead of animosity So the person that's probably going to Hate this most is the general contractor Or the subcontractor who's actually Doing the work but we've had these Discussions so imagine Um you've done a week's worth of of Really hard work you think everything is Perfect no one comes to work wanting to Do a bad job mistakes happen we're only Human would you rather know that that Mistake happened while you're still on Site and you haven't gone to another job Or do you want to know three months down The line when you've gone to another job You've got other responsibilities you Just want to go home to your family so Once you put it in in that respect then They understand the value of it and they See that in the end it's going to save Them time and money and energy Especially for instance one of the sites

We showed you Oslo when the data was Being captured there was minus 20 Degrees So nobody wants to go back in that Environment to do something and the fact That we can deliver that solution to Them right away so that they can fix it There's a huge amount of value another Thing is the entire operation of the Device if they're using our robot for Example It's poke you know you don't need to be A computer scientist to be able to Operate this thing it's a joystick a Remote you know a video game controller And they generally the best thing we've Seen is they're not in off the device They see it as a tool like a screwdriver Or a wrench on site this is another tool That helps them do their job better Thanks In the business model where you make Robots and then you have a SAS fee There's some payback associated with you Know getting recurring revenue and Upfront Hardware or investment like what What kind of math are you optimizing Around in terms of making the bits this Business model work So one of the things we're doing is We're optimizing around geography so We're focusing on a couple of locations Within Europe and the us at the moment To make sure that we can focus all our

Resources on a couple of projects that Are really delivering the most value for The solutions we have today but then We're also targeting companies that Already have existing reality capture Teams so they're already generating that Raw data that they just have no way of Really making use of so we're adding Value on top of an investment that They've already made and we're kind of Structuring it around a balance between Those two things and I would also kind Of mention that considering the value That we're delivering back to the Customers the cost of the robot is is Quite low in relationship to that Maybe off stage we can tell you how much It costs it's nothing Last questions All right one more round of applause for Scaled robotics Thank you great thank you so much thank You bye