When this caught my eye I got really interested. Google AI is launching a website titled rǝ which reconstructs cities from historical maps and photos. You might have seen the underlying tool last month but this productizes it a bit. What I find compelling about this effort is the output is a 3D city that you can navigate and review by going in back in time to see what a particular area looked like in the past.
Of course, Scottsdale, my town, is not worth attempting this on, but older cities that have seen a ton of change will give some great inside into how neighborhoods have changed over the past century.
Just take a look at the image above, it really does give the feel of New York back in the ’40s and earlier. People remember how a neighborhood looked, but recreating it in this method gives others key insights into how development has changed how certain areas of cities look and act.
This tool is probably more aimed at history professors and community activists, but as we grow cities into smarter, cleaner places to live, understanding the past is how we can hope to create a better future. I’d love to see these tools be incorporated into smart city planning efforts. The great part of all this is it is crowdsourced, open-sourced, and worth doing. I’m starting to take a deeper dive into the GitHub repository and look how the output of this project can help plan better cities.
On Monday I had a bit of a tweetstorm to get some thoughts on paper.
In there I laid out what I thought addressing inside a building should look like. A couple of responses came to the “why” or “this isn’t an issue” but the important thing here is with smart buildings, they need to be able to route people not only to offices for “business” but workers to IoT devices to act upon issues that might occur (like a water valve leaking in a utility closet). Sure one, could just pull out an as-built drawing and navigate, or in the case of visiting a company, the guard at the front door, but if things such as Apple Glass and Google Glass start becoming a real thing, we’ll need a true addressing system to get people where they need to be.
Apple and Google are working this out themselves inside their ecosystems but there needs to be an open standard that people can use inside their applications to share data. I mentioned Placekey as a good starting point with their what@where.
The what is an address – poi encoding and the where is based on Uber’s H3 system. As great as all this is, it doesn’t help us figure out where the leaky valve is in the utility closet. This all is much better than other systems and is a great way to get close. I’ve not seen any way to create extensions to Placekey to do this but we’ll punt the linking problem for now.
The other problem with addressing inside a building is the digital twin might not be in any projection that our maps understand. So we’ll need to create a custom grid to figure out where the IoT and other interesting features are located. But there seems to be a standard being created that solves just this problem, UBID.
UBID builds on the open-source grid reference system and is essentially the north axis-aligned “bounding box” of the building’s footprint represented as a centroid along with four cardinal extents.
I really like this, it might even compete with Placekey, but that’s not my battle, I’m more concerned with buildings in this use case. There is so much to UBID to digest and I encourage you to read the Github to learn more.
But if we can link these grids of buildings, with a Placekey, we have a superb method of navigating to a building POI and then drilling down into navigating to that location using all the great work that companies like Pixel8 are doing. But all that navigation stuff is not my battle, just a location of an IoT sensor in a digital twin that may or may not be in a project we can use.
Working toward that link, a unique grid of a digital twin to a Placekey would solve all problems with figuring out where an asset inside a building is and what is going on at that location. The ontologies to link this could open up whole new methods of interrogation of IoT devices and so much more. e911 and similar systems could greatly benefit from this as well.
The last time most people heard from Sidewalk Labs was when Toronto didn’t go forward with their Smart City project. There are a ton of reasons why that didn’t happy, but moonshots are what they are and even if you don’t reach the moon, outcomes can be really good for society. Of course, I know not what Sidewalk Labs has been working on but I have to assume Delve exists because of the work they are doing to build smart cities.
Delve at its most simple description is where computers figure out the best design options for commercial or residential project development. there is much more going on here and that’s where the Machine Learning (ML) part comes in and what really catches my eye. I’ve done a tone of work with planning in my years of working with AECs and coming up with multiple design options is time-consuming and very difficult. But with Delve, this can happen quickly and repeatable in minutes.
You get optimal design options based on ranked priority outcomes such as cost or view. Delve takes inputs such as zoning constraints (how high a building can be or what the setbacks are), gross floor area (commercial or residential), and then combines these with the priority outcomes. Then you get scored options that you can look further into and continue to make changes to the inputs.
The immediacy of this is what really sets this apart. When I was at Cityzenith years ago, we attempted to try and get this worked out but the ML tools were not developed enough yet. Clearly though, with Alphabet backing, Sidewalk Labs has created an amazing tool that will probably change how cities are being developed.
I am really excited to see how this works out. I don’t see an API yet so integration outside of Sidewalk labs does not seem to be a priority at this point but the outcome for scaleable planning like this needs to have an API. I’ll be paying attention but seeing ML being used for this type of development is logical, understandable, and workable. We should see great success. You can read more at the Sidewalk Labs blog.
Digital Twins are easy. All you have to do is create a 3D object. Some triangles and you’re done. A BIM model is practically a Digital Twin. The problem is usually those twins are created from data that isn’t “as-built“. What you end up with is a digital object that ISN’T a twin. How can you connect your IoT and other assets to a 3D object that isn’t representative of the real world?
I talked a little bit last time on how to programmatically create digital twins from satellite and other imagery. Of course, a good constellation can make these twins very up to date and accurate but it can miss the details needed for a good twin and it sure as heck can’t get inside a building to update any changes there. What we’re looking for here is a feedback loop, from design to construction to digital twin.
There are a lot of companies that can help with this process so I won’t go into detail there, but what is needed is the acknowledgment that investment is needed to make sure those digital twins are updated, not only is the building being delivered but an accurate BIM model that can be used as a digital twin. Construction firms usually don’t get the money to update these BIM models so they are used as a reference at the beginning, but change orders rarely get pushed back to the original BIM models provided by the architects. That said there are many methods that can be used to close this loop.
Companies such as Pixel8 that I talked about last week can use high-resolution imagery and drones to create a point cloud that can be used to verify not only changes are being made as specifications but also can notify where deviations have been made from the BIM model. This is big because humans can only see so much on a building, and with a large model, it is virtually impossible for people to detect change. But using machine learning and point clouds, change detection is actually very simple and can highlight where accepted modifications have been made to the architectural drawings or where things have gone wrong.
The key point here is using ML to discover and update digital twins at scale is critically important, but just as important is the ability to use ML to discover and update digital twins as they are built, rather than something that came from paper space.
Let’s face it, digital twins make sense and there is no arguing their purpose. At least with the urban landscape though, it is very difficult to scale digital twins beyond a section of a city. At Cityzenith we attempted to overcome this need to have 3D buildings all over the world and used a 3rd party service that basically extruded OSM building footprints where they existed. You see this in the Microsoft Flight Simulator worlds that people have been sharing, it looks pretty good from a distance but up close it becomes clear that building footprints are a horrible way to represent a digital twin of the built environment because they are so inaccurate. Every building is not a rectangle and it becomes impossible to perform any analysis on them because they can be off upwards of 300% on their real-world structure.
How do you scale this problem, creating 3D buildings for EVERYWHERE? Even Google was unable to do this, they tried to get people to create accurate 3D buildings with Sketchup but that failed, and they tossed the product over to Trimble where it has gotten back to its roots in the AEC space. If Google can’t do it who can?
Vricon, who was a JV between Maxar and Saab but recently absorbed by Maxar completely, gives a picture into how this can be done. Being able to identify buildings, extract their shape, drape imagery over them, and then continue to monitor change over the years as additions, renovations, and even rooftop changes are identified. There is no other way I can see that we can have worldwide digital twins other than using satellite imagery.
Companies such as Pixel8 also play a part in this. I’ve already talked about how this can be accomplished on my blog; I encourage you to take a quick read on it. The combination of satellite digital twins to cover the world and then using products such as Pixel8 can create that highly detailed ground truth that is needed in urban areas. In the end, you get an up to date, highly accurate 3D model that actually allows detailed analysis of impacts from new buildings or other disruptive changes in cities.
But to scale out a complete digital twin of the world at scale, the only way to accomplish this is through satellite imagery. Maxar and others are already using ML to find buildings and discover if they have changed over time. Coupled with the technology that Vricon brings inside Maxar, I can see them really jump-starting a service of worldwide digital twins. Imagine being able to bring accurate building models into your analysis or products that not only are hyper-accurate compared to extruded footprints but are updated regularly based on the satellite imagery collected.
That sounds like the perfect world, Digital Twins as a Service.
When you think about IoT you think about little devices everywhere doing their own thing. From Nest thermostats and Ring doorbells to Honeywell environmental controls and Thales biometrics; you imagine hardware. Sure, there is the “I” part of IoT that conveys some sort of digital aspect, but we imagine the “things” part. But the simple truth of IoT is the hardware is a commodity and the true power in IoT is in the “I” part or the messaging.
IoT messages are usually HTTP, WebSockets, and MQTT or some derivative of them. MQTT is the one that I’m always most interested, but anything works which is what is so great about IoT as a service. MQTT is leveraged greatly by AWS IoT and Azure IoT and both services work so well at messaging that you can use either in replacement of something like RabbitMQ, which my daughter loves because of the rabbit icon. I could write a whole post on MQTT but we’ll leave that for another day.
IoT itself is built upon this messaging. That individual hardware devices have UIDs (unique identifiers) that by their very nature allow them to be unique. The packets of information that are sent back and forth between the device and the host are short messages that require no human interaction.
The best part of this is that you don’t need to hardware for IoT. Everything that you want to interact with should be an IoT message, no matter if it is an email, data query or text message. Looking at IoT as more than just hardware opens connectivity opportunities that had been much harder in the past.
Digital twins require connectivity to work. A digital twin without messaging is just a hollow shell, it might as well be a PDF or a JPG. But connecting all the infrastructure of the real world up to a digital twin replicates the real world in a virtual environment. Networks collect data and store it in databases all over the place, sometimes these are SQL-based such as Postgres or Oracle and other times they are simple as SQLite or flat file text files. But data should be treated as messages back and forth between clients.
When I look at the world, I see messaging opportunities, how we connect devices between each other. Seeing the world this way allows new ways to bring in data to Digital Twins, think of GIS services being IoT devices, and much easier ways to get more out of your digital investments.
The GIS world has no idea how hard it is to work with data in the digital twin/BIM world. Most GIS formats are open, or at works readable to import into a closed system. But in the digital twin/BIM space, there is too many close data sets that makes it so hard to work with the data. The loops one must go through to import a Revit model are legendary and mostly are how you get your data into IFC without giving up all the intelligence. At Cityzenith, we were able to work with tons of open formats, but dealing with Revit and other closed formats was very difficult to the point it required a team in India to handle the conversions.
All the above is maddening because if there is one thing a digital twin should do, is be able to talk with as many other systems as possible. IoT messages, GIS datasets, APIs galore and good old fashioned CAD systems. That’s why open source data formats are best, those that are understood and can be extended in any way someone needs. One of the biggest formats that we worked with was glTF. It is widely supported these days but it really isn’t a great format for BIM models or other digital twin layers because it is more of a visual format than a data storage model. Think of it similar to a JPEG, great for final products, but you don’t want to work with it for your production data.
IFC, which I mentioned before, is basically a open BIM standard. IFC is actually a great format for BIM, but companies such as Autodesk don’t do a great job supporting it, it becomes more of interchange file, except where governments require it’s use. I also dislike the format because it is unwieldy, but it does a great job of interoperability and is well supported by many platforms.
IFC and GLTF are great, but they harken back to older format structures. They don’t take advantage of modern cloud based systems. I’ve been looking at DTDL (Digital Twins Definition Language) from Microsoft. What I do like about DLDT is that it is based on JSON-LD so many of those IoT services you are already working with take advantage of it. Microsoft’s Digital Twin platform was slow to take off but many companies, including Bentley Systems, are leveraging it to help their customers get a cloud based open platform which is what they all want. Plus you can use services such as Azure Functions (very underrated service IMO) to work with your data once it is in there.
The magic of digital twins is when you can connect messaging (IoT) services to your digital models. That’s the holy grail, have the real world connected to the digital world. Sadly, most BIM and digital twin systems aren’t open enough and require custom conversion work or custom coding to enable even simple integration with SAP, Salesforce or MAXIMO. That’s why these newer formats, based mostly on JSON, seem to fit the bill and we will see exponential growth in their use.
I mean I could go on, but Studio seems to be the new method of naming authoring tools. I’m not here to make fun of this, just state that I love the name studio used in this sense. Having worked with Architects most of my life, Studio has a great connotation for workspace. All those apps above are used as a workspace to create something else. The concept of a studio, used this way, really helps define what a product is used for. I think the term, hackspace, has taken on a modern connotation for studio but the core concept is just so sound on this part.
Pro or Professional probably harkens back to the early days of software and hardware, where you’d create consumer and professional products. These days, professional is usually used to show a higher end product, vs a professional product (or at least used inconsistently). But appending “pro” to an end of a product name doesn’t signify the same purpose as studio does.
One could almost call ArcMap or QGIS a studio since that is what people are doing with it. My personal studio is my office with this computer I’m writing this blog post on. That thought makes me smile.
Happy Friday everyone. These weeks just fly by when you are locked in your house looking out your front window for the Instacart delivery from the grocery store. I just wanted to remind everyone that I’ve got a weekly newsletter were I do some deep dives into things that are on my mind related to GIS, BIM, Smart Cities and technology.
Just sign up below and you’ll get a newsletter in your inbox each Wednesday (or maybe Thursday LOL).
Please check your email to confirm your newsletter subscription.
I’ve spent years trying to build worldwide building datasets for Smart City and Digital Twin applications. I’ve tried building them using off-the-shelf data providers that give you COLLADA files, I’ve tried using APIs such as the Mapbox Unity SDK and buying buildings one by one to fill in gaps. None of these solutions have the resolution needed to perform the types of analysis needed to make better choices for cities and development potential. How to create real 3D cities with enough resolution has been out of our grasp until now.
I’ve been following Pixel8 for a while now and it is clear that crowdsourcing these models is going to be the only way forward. Over 10 years ago, Microsoft actually had this figured out with their Photosyth tool but they never were able to figure out what to do with it. Only today are we seeing startups attack this problem with a solution that has enough resolution and speed that we can start seeing cities build highly detailed 3D models that have actual value.
It is still early days with these point cloud tools, but at the speed they’ve improved over the last year, we should be seeing their use more and more. Mixing the data from smartphones, lidar and satellite imagery can make large areas of cities mapped in 3D with high accuracy. Pixel8 isn’t the only company attempting this so we should see real innovation over the next year. Stay tuned!