Talking to Boston Dynamics' Brian Ringley about the future of construction robotics and automation
Meet one of the most important people working in construction robotics today. Brian Ringley, Construction Technology Manager at Boston Dynamics, talks with Constructible about Spot's current and future use cases, how innovative contractors use Spot today, 'digital twin starter sets', what construction bots might look like in 2040, upskilling for an automated future, and more.
13 questions for Boston Dynamics' Brian Ringley about Spot the Robot Dog and the future of robotics in construction.
Tell us a bit about your background in construction and automation.
I graduated in the recession and could not get a professional job in architecture, so I ended up working in digital fabrication labs, on a design technology team, teaching in academia, and picking up a lot of technology skills along the way.
While I was a construction automation researcher at WeWork, we asked our superintendents and assistant supers, "If we could automate one thing you did to make your life easier, what would you like us to work on?" They said, "Please do my daily photo documentation or job walk."
During this work, we found drones to be unfeasible for interior commercial construction spaces in terms of battery life, payload capacity, and operational safety. We tried wheeled and tracked robots, both those that we developed ourselves with kits, and ones that were available on the market. Ultimately, there was no method of wheeled or tracked locomotion that could get everywhere on the site that we needed it to get.
In 2018, Boston Dynamics released a video of Spot on their Japanese customer construction sites and I said, "We've really got to try this." We became Spot early adopters and were able to get the robots on our sites, and demonstrate that they could get anywhere in the environment. It could autonomously trigger 360 image capture and it could do it repeatedly and autonomously.
Most importantly, the Spot software development kit (SDK) allowed us to upload that data to our construction project tracking software over jobsite WiFi. When I saw the power of Spot’s legs to get anywhere on the job site where a sensor is needed, it became very clear to me that Boston Dynamics was the company where I could have the most impact on the construction industry.
Brian Ringley, Boston Dynamics, Jason Fields, Mortenson, and Derek Smith, Mortensen, talk about using Spot the Robot Dog integrated with Trimble solutions on the jobsite.
What does your role as a construction technology manager at Boston Dynamics involve?
I'm a product manager for Spot, with a specific focus on the construction industry. I work very closely with our construction customers — typically general contractors as well as allied industry customers, architects, engineers, and real estate professionals, to make sure that we are setting good ROI hypotheses for the use of Spot in their environments. We use feedback and signals from those customers to inform new features for Spot as we continuously update the software.
I also work with our developer partners to make sure that we are pushing the Spot SDK in ways that make it easier for them to add value on top of the robot.
What is the biggest benefit Spot the Robot Dog provides to contractors?
The most obvious value Spot offers right now is to general contractors because they are responsible for the timely, on budget delivery of their projects. With Spot, GCs have the opportunity to establish a rigorous, repeatable data capture program, while decreasing the labor and resources required to do so. That's a huge win for them.
For example, if I'm a construction project manager and I'm really struggling to do data capture or continuous scanning of my site in order to automate work-in-place quantification, I don't want to have to go separately to a robotics company and a software company and a hardware company and to be required to figure out the various SDKs and APIs — I might not know what those acronyms even mean — in order to pull all of that technology together.
Instead, I want to go to Trimble and say, 'I need a scanning robot,' or 'I need a mobile scanning solution,' and that happens to be attached to this amazing robot, but it's solving my business need and that's really the point.
How does Spot help upstream and downstream stakeholders, such as architects or real estate management?
The value upstream to, say, the architect or engineer (something I've discussed in a piece recently in Architect Magazine) is first and foremost to perform better construction administration. You have a few architects designated for that role, while the majority of the designers are sitting in the office at their computers. With robotic telepresence or teleoperation, the designers and upstream professionals can actually have an onsite presence. This creates the freedom and flexibility for superior coordination, and the ability to catch errors, reconcile discrepancies, and deal with gaps in documentation.
Another way a robot like Spot helps designers, GC’s and ultimately, owners and operators, is with better delivery of an as-built. If we can combine some of the elements of construction progress over time into an architect’s Building Information Model (BIM), and carry that data all the way through to owner handoff, this will set the stage for the owner to have access to historical asset data, which is just as important as the 3D geometry within the model.
I think that the GC and the architect and the engineer and all of these stakeholders that deliver the building have to figure out how to put together a set of services that essentially forms the foundation for the digital twin. Then the owner can take that package and run with it — integrating all sorts of other real-time data systems. Imagine now that contractors and designers can actually continue to have a relationship with the owner over the full building lifecycle, and continue to earn revenue, by supporting the owner's setup and operation and of the digital twin.
Spot plays a pivotal role here because mobile automation or automation that moves through the site is how GC’s are able to stream real-time data for construction progress. You can't really rely on fixed systems in a dynamic and difficult-to-predict environment like a construction site.
A construction site is temporary, it's changing every day, it's chaotic. Owners typically don't want to invest in a lot of sensor infrastructure at that stage. Having the mobility of Spot really works well at that phase. After the construction phase, Spot gives contractors valuable data to turn over to the owner, who could continue to use Spot for things like routine maintenance monitoring.
We see customers like Reply and ECE in the real estate space showing how Spot is a building lifecycle tool, really setting the owner up for success by connecting to digital smart building systems. Whether these systems are for HVAC or tenant security or parking management, there are all sorts of things that an owner would want to do once the building is in operation, and that all relies on streaming data.
Fast forward 20 years from now — where do you think we’ll be in terms of construction automation and robotics?
It's pretty extraordinary to think about where we might be in 20 years, given where we were 20 years ago. The first quadrupeds by Boston Dynamics hadn't even left the lab 20 years ago. It can be hard to predict where we're going from here but I think we'll see a proliferation of many types of robots from general mobility platforms to single-task robots.
We’ll see more task-specific robots that do one construction trade labor task very well, like a drilling robot or a painting robot. 5G and future generation networks on connected job sites will make it possible for all of these different robots to collaborate and exchange data. Think of these as heterogeneous robot fleets — different types of robots specializing in different functions but also capable of being networked together to provide for two major advantages. First, we can consolidate operator control over the job site in the same way that you'd use a PLC system for industrial robotics in a factory or a warehouse. Second, these fleets will complete more complex tasks in concert than would be possible with any single robot individually.
Legged robots will continue to push valuable automation into human-purposed environments and dangerous or hard to access areas, or whenever you need an agile mobility solution for a relatively heavy payload like a laser scanner. Aerial drones, which are already fairly well-adopted by the industry, will continue to evolve to be even more valuable and sophisticated, and will intelligently interface with robots like Spot for data transfer and recharging.
I think we'll see specific developments for vertical projects versus horizontal projects. For high-rise construction we are already seeing automated cranes and scaffolding systems, and in the future we will see building envelope and envelope retrofit installation systems.
One of the more exciting side effects of getting all of these robots on site is job creation in terms of data analysts, data scientists, and other types of engineers, who will be able to create systems that automate reporting and insights out of the data the robots collect. We’re going to need a lot more software smarts to fully leverage the advantages of physical automation in construction.
The Construction Robots are Coming by The B1M.
As the building process becomes more automated, will the design of our buildings change? In 2050, will our cities look wildly different?
I'm a bit of a market realist when it comes to architectural styles and aesthetics (and a recovering architect besides). I don't think that technology actually is the primary driver of what buildings look like. Design will remain much more steered, at least in cities, by your lot line and how you can maximize it.
I think technology is more important in buildings when it comes to the comfort and health of their occupants — making sure there’s functioning adequate filtration for COVID, for example, or making sure that people are thermally comfortable while also conserving energy.
I’m interested in how we can take what we've learned from hundreds of years of modern commercial construction techniques, and continue to make those even more efficient so that we can drive down the cost of real estate, making structures more affordable, more efficient, sustainable, and longer-lasting.
There's so much that technology can do that doesn't impact the form of a building. For example, the high-rise development around Barclays Center in Brooklyn that is the tallest modular construction project in the world. It doesn't look particularly different from any other tower in Brooklyn, it was just constructed more efficiently. We often overemphasize aesthetics and form, when that tends to miss the point of how we can improve the built environment to serve the public.
The tallest modular construction project in the world, Brooklyn's Barclays Center in New York City, USA. Image by Dariusz Gryczka via Shutterstock.
Tell us about a current Boston Dynamics construction technology project that has you excited.
We have a joint customer with Trimble, where they've been able to run Spot missions in some of their tunneling projects. Tunnels and underground environment are a robotics challenge in terms of the perception systems that drive autonomous behavior, as well as the communication systems that allow you to monitor that autonomy and stay connected to the robot.
What's exciting about this use case is that there's a lot of investment in global railway infrastructure. This often requires sending robots through tunnels and the fact that Spot is able to continuously scan these environments really shows a path forward, not just for construction companies that are doing tunneling infrastructure, but also for the mining industry. Mining has historically heavily automated its processes and is always looking for additional advantages of automation. So I’m pretty excited about that one.
Another use case I'm excited about is the use of Spot to monitor renewable energy infrastructure, specifically solar farms. There is an extraordinary amount of this type of development over the next decade, so it represents a key opportunity to scale the use of robots in construction. Trimble and Boston Dynamics supported Mortenson's civil group in a solar farm exercise last year and while there is still work to do to make this an out-of-the-box solution, the pathway to providing value in these environments is very clear.
Are there any other companies doing really cool things with Spot?
There are a lot of interesting customer stories and we try to tell as many of them as we can on our website. A really technically savvy one that comes to mind immediately is what Brasfield & Gorrie has done with the Spot SDK, where their virtual design and construction (VDC) team and innovation team designed custom hardware for the robot, attached it, and wrote their own software to manage mission creation for data capture. They were also the driving customer force behind the integration of Spot into DroneDeploy's 360 walkthrough feature which was released last year.
We also have customers that are really savvy in a business sense. We worked with another contractor, Swinerton, and they were looking to solve a very specific business problem: They weren't paying their subs fast enough. Their goal was to automate their payment systems and get that payment time down.
This incentivizes the subs to bid more competitively, knowing that they’ll get that guaranteed quick payment. It also might incentivize some additional fees from the owner, for running a tighter ship and guaranteeing that the books are up to date. This would be a highly differentiating factor in the selection of a GC by an owner. Swinerton realized that in order to automate payments, they first needed some trusted mechanism for quantifying work-in-place — and that mechanism needed to be automated too.
That requires data about what's been done. In turn, that leads to the question, 'How do we establish a rigorous data capture program, preferably one that's very low labor, semi-autonomous?' That is what ultimately led them think of Boston Dynamics and perform their trialing of Spot.
I love this story because they adopted Spot as part of a solution to a complex business problem. There's a lot of value in companies that are curious about robotics and exploring new use cases, but I also find it interesting and helpful when Spot is derived as a critical part of the answer to a current business problem.
What’s your opinion on swarm bots? Fantastical and fringe or well-suited for realistic construction application scenarios?
That's an interesting question. Swarm robotics implies that higher-level intelligence comes in the aggregate of many robots, but each individual robot has a limited knowledge of the problem space and is working on a smaller piece of that. The intelligence of all of those robots combined together can achieve something complex.
One could imagine swarms lending quite a bit of efficiency to data capture with robots of limited battery life, groups of drones for instance. We know that drone swarms are a real technology used largely in the entertainment space to do light shows and concerts, but I can imagine a group of drones scanning or photographing a site. Rather than relying on a single flight path, they would intelligently cover much more space.
There are many opportunities for things like fleets of Spots or other robots to work together, to achieve more complex tasks than would be possible with a single robot. Hauling something heavy, like material conveyance on site is one possibility. You might have one robot that is doing a particular task that requires materials, and then other robots actually transporting those materials required for the task. One manipulates materials, while the other one conveys the materials. We're just starting to realize the potential here, but even with this robotic functionality, operators would need better software for robot swarm and fleet management for this to be valuable.
Like I mentioned before, some simple examples are pulling something heavy. One of our earliest videos was a bunch of Spots pulling large truck.
A pack of Spot robots pulls a large truck.
How can people working in construction now upskill in a way that will help them capitalize on robotics and automation in the future?
At Boston Dynamics, we're trying to make it so that this is a frictionless experience for construction professionals. We have a lot of respect for the way they do work right now. We're trying to give them another tool in their tool belt to do their job more efficiently.
I don't think it's a bad thing for a construction professional to invest in learning the basics of coding and Python. That's a really powerful skill. That's just useful in so many aspects of one's life today. It’s ridiculous for me to say ‘Oh hey, everyone in construction needs to be a computer scientist’, although you will here that from some contingents
Robots are tools to make people more productive, or to remove them from harm's way. The onus is on us, as in Boston Dynamics and other companies that are bringing robots into the world, to build robots that are easy and intuitive and safe to use — robots that have an obvious purpose and an expected set of behaviors for industry professionals.
For the time being, as a customer of this new technology, it's less about working physically with the robots or programming new robot behavior, and more about understanding how to create value from the data the robot is creating. We’re in a completely unprecedented era of data capture for construction professionals. I’ve mentioned some of the possibilities ranging from automated work-in-place analysis to the handoff of digital twins to owners — those opportunities are there for enterprising professionals to capitalize on.
So maybe it’s not exactly about upskilling, but more about construction professionals and companies investing in data analysis — so companies like Boston Dynamics can come up with more and more useful applications for Spot?
Yes, it's sort of along those lines. I've always loved the idea of empowering entrepreneurial project managers and supers and other people who are managing these sites, to be the ones who lead the innovation efforts because they're the ones experiencing the difficulties firsthand and would benefit the most from increased efficiency.
I think a cool idea would be to reward project managers and supers with extra budget (or even personal bonuses) in order to drive more grassroot innovation. I would love to see more investment and empowerment of these professionals who say, ‘Give me a robot. Give me this piece of software, give me this camera, this scanner, this technology. I intuitively know how this could make my life easier and can provide feedback quickly.’
Something for construction companies to think about, again, in an era of unprecedented technology development is: How do innovation groups keep up with it? How do they manage innovation? I can see the struggle under the weight of so many new technologies and solutions, so how do you know which ones are really making a dent in your margins? I think we'll start to see more innovation in how we innovate.
Brian packs up Spot after a day in the field. He loves to see construction project managers and superintendents intuitively use the robot to make their lives easier and appreciates their unique ability to provide quick, practical feedback and breakthroughs in use cases.
Outside of construction technology is there anything that helps you think differently about your job and the future of robots?
As a former architect, still take inspiration from architecture and designers like Jeanne Gang, Amanda Levete, and Elizabeth Diller, who are really at the top of their respective games right now. I also do quite a bit of reading and am inspired by philosophers thoughtfully considering our rapidly changing world like like Donna Haraway and Paul Virilio.
I like to read a lot of sci-fi. I think there are deeply interesting ideas from the works of Ted Chiang (who actually wrote the story that inspired the movie Arrival) and Jeff VanderMeer, who is best known for his novel Annihilation. And sci-fi films, too — I'm really interested in the 1970s stuff. That was a really interesting era where there were thought-provoking narratives combined with practical effects that made for visually compelling and thought-provoking films like The Andromeda Strain, THX 1138, Solaris, World on a Wire. Even some of the Godzilla films from that era (Mechagodzilla!) are pretty smart in their own weird ways.
What are some resources you can recommend for readers who want to dive deeper into construction automation and robotics?
Aside from all the conversations on social media of course, for AI I recommend Judea Pearl for thinking about causality, and Stuart Russell for thinking about how to make AI provably beneficial to humanity. I wouldn't call Russell a fearmonger, but he definitely points out the ways in which having systems smarter than us might not be so great, and how it's never too early (regardless of how advanced you consider that technology to be) to think about how to design for that.
For construction robotics, there are a lot of really interesting books out there, but I think my favorite one is a book from the '90s, called Construction Robots: The Search for New Building Technology in Japan, which is about the efforts of the large Japanese contractors in the 1980s to develop and build their own task-specific construction robots and construction automation systems. It's really miraculous how advanced some of these robots were in the '80s. It's also a cool way to learn about the construction industry in Japan, which is arguably the most innovative construction industry in the world.
For industrialized construction and pre-fab stuff, which I think is extremely important as well, I really like this book called Refabricating Architecture by Kieran Timberlake, which is from an architect's perspective. It's almost 20 years old now but I still find it to be one of the most relevant books in terms of design thinking around new methods of construction.
I also listen to a lot of podcasts. Some of my current favorites are The Robot Report for general robotics stuff. For construction, I like The Construction Brothers, The ConTechCrew. For the changing architecture industry and its implications on design careers, I enjoy TRXL and Practice Disrupted. I am a little late to it but I just learned about Space To Build. That one specifically focuses on women who are innovating and working in the construction industry. That's another one that's just full of some really fantastic guests.
** This article was originally created by Rachel Jones for constructible.trimble.com