Fastbrick’s Hadrian X prototype.

RobotLabNYC’s third installment will be on June 13, in New York City with Howard Morgan (FirstRound Capital) and Tom Ryden (MassRobotics); together, we will be “Exploring The Autonomous Future” (RSVP today). Coincidentally, Jimmy Fallon featured a new bit this week called “Showbotics,” providing viewers with a sneak peek into the robotic future:

While Fallon pokes fun, the reality is that robots today are showing up for work in record numbers. As America pulls out of NAFTA and starts a trade war with Canada over lumber imports, it is predicted that home building costs could increase by more than 20% over the next year. In order to keep America building without sacrificing margin, labor is shifting from humans with tool belts to job-ready robots.

An example of machines being added to the field is MIT’s Digital Construction Platform (DCP) – a 3D-printing fabrication robot. The DCP consists of a robotic arm that can print structures made of insulating foam at architectural scale — in other words, big enough to live in. In a test last July, MIT researchers printed an open dome structure in two days that was as large as a ranch house. The average single-family home takes about 7 months to build (US Census Bureau).

According to the white paper published this week in Science Robotics, the researchers claim, “this printed test structure is currently both the largest monolithic structure ever 3D-printed by an on-site mobile platform and the fastest autonomously printed architectural-scale structure.”

Shanghai-based WinSun 3D-printed walls last year for a six-story apartment building out of concrete that was assembled on site for immediate occupancy. According to WinSun, they were able “to save 60% of the materials typically needed to construct a home and can be printed in a time span which equates to just 30% of that of traditional construction. In total, 80% less labor is needed, meaning more affordable construction, and less risk of injury to contractors.”

In the USA, Blueprint Robotics is manufacturing wood-framed houses today with industrial robots. Builders upload their architectural plans to Blueprint’s cloud-based platform which outputs within weeks pre-made walls, floors, and roofs. While the concept of pre-fabricated houses is not novel -Sears Roebuck & Co first introduced kits to homesteaders in the 1900s – Blueprint is the first to create an automated production line that pre-places everything from sheetrock to nails to lighting placement. This means modular structures can now be used for bigger and more luxurious developments from multi-million dollar mansions to large apartment buildings. Bloomberg News reported this week that even Marriott, the world’s largest hotel operator, is “turning to modular construction for some of its properties,” shipping the walls inside the same containers as the television sets and beds.

According to leading home consultant, John Burns, “this has to be the wave of the future — I don’t know how we solve the labor shortage otherwise. What drives modular construction is the ability to build the house more cost-effectively.” The Bureau of Labor states that almost 200,000 construction jobs are unfilled across the country at a boom time for homebuilders. The U.S. construction industry’s unemployment is at its lowest level in a decade as fewer young people enter the construction workforce. The door remains wide open for robotic disruption.

Jerry Smalley, CEO of Blueprint, claims his mechanized platform is filling this labor gap. While manual labor is greatly reduced, skilled tradesmen are still high demand, according to Smalley, “robots cut the hole, but somebody still has to put the electrical box and pipes in the right places.” Blueprint utilizes Germany-based Weinmann’s Stiles machines that have been shipping pre-fabricated homes in Europe for decades.

Construction globally generates over $8 trillion annually. It is also one of the most labor-intensive professions with concrete casting, wood framing and bricklaying. As a result, the construction industry accounts for almost 20% of US workplace fatalities. Masons are particularly prone to musculoskeletal injuries, due to repetitive motion strain from carrying and laying over 500 bricks a day by hand.

New York-based Construction Robotics is revolutionizing the masonry trade with its SAM100 (Semi-Automated Mason) robot, capable of working six times faster than the average human. SAM uses a combination of a conveyor belt, robotic arm, and concrete pump. Construction Robotics recently upgraded its product to enable the laying of popular “solider course” bricks. SAM costs the equivalent of the average salary of 10 masons or $500,000 and is already being deployed on sites throughout America.

Construction Robotics’s head of business development, Zachary Podkaminer said, “we don’t see construction sites being fully automated for decades, if not centuries. This is about collaboration between human workers and machines. What SAM does is to pick up the bricks, put mortar on them, and puts it on the wall. It still requires a mason to work alongside it. SAM’s just there to do the heavy lifting.”

Already Construction Robotics has competition from Australian-based Fastbrick Robotics. Fastbrick’s Hadrian X prototype claims to place bricks at an impressive rate of 1,000 per hour. The robot (shown above) can handle different brick sizes as well as cut, grind, mill, and route the bricks to fit any structure before putting them in place. Unlike SAM, which uses traditional bricks and mortar, the Hadrian machine is optimized for interlocking precision bricks (that are 15 times larger) and construction adhesive to hold the concrete structure.

Fastbrick Operation Manager, Gary Paull, says his system “will produce structurally sound wall to renderable quality at speeds never seen before.” The company plans to market the Hadrian bricklayer first in Western Australia and then to the rest of the country before offering it worldwide.

Deployment of task-orientated robots will be advanced further by the coming inevitability of autonomous vehicles ferrying heavy construction equipment and materials on the site.  According to Jenny Elfsberg of emerging technologies at Volvo CE, “Autonomous machines will increase safety in hazardous working environments and eliminate the possibility of accidents caused by human error. They will also perform repetitive tasks more efficiently and precisely than a human operator. And because machines will be operated in the most efficient way, customers will benefit from improved performance, productivity, fuel efficiency and durability.”

In a one-hour comparison, it was found that the autonomous wheel loader could reach the equivalent of 70% of that of a skilled operator’s productivity levels when loading and unloading. The machine has also done ‘real work’ for a Volvo CE customer at an asphalt plant in Sweden.

“In the future, you could also potentially have one operator for three or four machines, increasing productivity and further decreasing costs,” Elfsberg states. “Looking ahead, I imagine that autonomous machines will be smaller and more robust. There will be no need for a cab or suspension – much like the HX1 concept [shown above],” which Volvo CE unveiled last year.

Mining company Rio Tinto is deploying a fleet of close to 100 driverless dump trucks (manufactured by Komatsu) to move high-grade ore autonomously. It is also developing and testing an autonomous heavy-haul long-distance railway system, and has deployed an automated blast-hole drill system that allows a single operator to remotely control multiple drill rigs.

According to a Rio Tinto press release, “these driverless vehicles deliver their loads more efficiently, minimizing delays and fuel use, and are controlled remotely by operators who exert more control over their environment and ensure greater operational safety.”

Industry leaders, like Caterpillar and GE, are taking note by funding innovation at early financing rounds that are typical for corporate venture. Last year GE and Caterpillar invested in Clearpath Robotics’ $30 million Series B round to expand its fleet of autonomous mobile robots into industrial settings, like construction and mining. Previously, Clearpath’s mobile robots were primarily utilized in warehouses moving boxes and pallets around distribution centers.

This past March, Caterpillar co-hosted an industrial startup competition at SxSw’s Interactive Conference with their local dealer Holt. Leading autonomous navigation company 5D Robotics walked away with the $20,000 grand prize. 5D is currently working with United Rentals to convert their fleet of manual industrial machines into autonomous mobile robots. While Caterpillar has been working on self-driving trucks internally since the 1990s, reaching outside for innovation could be the difference in dominating the new autonomous landscape.

This new sentiment was reflected in its “Year In Review” investor statement: “For 90 years, Caterpillar has delivered breakthrough innovation inside our machines and engines. Today, that innovation is increasingly happening outside the machine. We’re going ‘beyond the yellow iron,’ harnessing the power of big data to offer our customers insights that decrease operating costs, increase uptime and maximize profitability.”

Launched in 2009 by a group of Waterloo engineering students, Clearpath‘s unmistakeable bright yellow and black robots have become synonymous with unmanned vehicle research in university research labs around the world. Now, as the field of robotics matures, Clearpath is forging into industrial applications, too. We caught up with CTO Ryan Gariepy at the 2015 Field and Service Robotics conference (FSR), to talk about their roots in research, the role of ROS and open source in their business model, and the challenges and opportunities of launching a robotics startup in Canada.

[HS] You got your start in the business of robotics while you were still a student at Waterloo. What made you decide to launch a robotics company at that time? Did you have a specific project or technology you were developing that made you think, “Hey, this could actually be useful out there!” …? Or was it that you saw the potential in the field more generally?

[RG] We started Clearpath right out of university in 2009 as we finished our undergraduate degrees – I did my Masters at the same time as we were getting off the ground, which is generally regarded to be a bad idea in terms of stress levels. It’s hard to believe that we just celebrated our sixth anniversary this year.

Starting the company was less about the technology and more about the potential … we just looked at everything around us and said, “Why do we have to shovel snow? This can be automated. Why do we have to do any work that does not require human creativity?” There are a lot of dangerous jobs out there that most people don’t realize are performed on a daily basis by tens of thousands of people in order to keep society going, like mining, farming, construction, stocking warehouses. And yet we need fuel and food and buildings, and we want our same-day deliveries from the Internet. There is a demand for this kind of work to take place.

Around the time we were starting out, Google Maps was getting off the ground. The concept of having information available at your fingertips was also becoming compelling and valuable to people, and the best way to get information about the world on a constant basis is to involve robotics or automation in some form. At the same time, sensors and the cost of sensing were improving due to the rapid uptake in smart phones; we were seeing advances in laptops, smartphones, and electric and hybrid vehicles driving down costs and improving battery density.

People now accept that robotics, automation, and intelligent systems can actually solve some of their problems.

All of this was coming together and we felt that autonomous robotics was possible at a relatively high reliability and low cost, especially when compared with what had been done before.

What would you say have been the biggest changes that you’ve seen in the field of robotics since then?

I think one of the biggest changes is simply that people now accept that robotics, automation, and intelligent systems can actually solve some of their problems.

This recognition means that there are a lot of opportunities out there, both in terms of hardware and software. In some industries – ones that have been around for a long time – they already know how to build solid hardware that could be made robotic, but they need the right parts and the right software to automate it. Then there are other industries that need full solutions. There are a lot of pieces starting to come together now.

Your early customer base was made up of mainly universities and research labs. Do you find yourself getting pulled into a different direction now that you’re expanding more into industry?

The research market is definitely maturing and our product development cycle within that side of the business is very well understood. We’ve released two completely different platforms in the last eight months alone.

Now we are in a position where we can leverage the experience we gained from the R&D market to address challenges in industry. We are actively pushing the development of full robotic solutions (including software), and will soon be releasing a new platform along these lines.

How do you manage the problem of building in more than one direction at the same time?

One of our biggest challenges is that there is no one single clear challenge, especially at the scale we’re at right now. At this stage there are many things that need to be focused on: sales, operations, quality assurance, new features, understanding where the market’s going, product management, marketing … so the challenge is more about managing the priorities of any given department.

How do I provide guidance to the team so that they can spend their time on the right tasks? How do we remain dynamic as a startup? How do we remain flexible enough to change direction quickly if we need to, but still be stable enough to stick to our core values? These are the kinds of questions we are dealing with.

Speaking of core values, Clearpath took a chance from the beginning by focusing on open source and ROS. Does open source make for a good business model?

Why open source? No matter how large a company we become, there will always be more smart people who can build and program robots outside our company than within it.

One of the reasons we looked at open source was the philosophy that no matter how large a company we become, there will always be more smart people who can build and program robots outside our company than within it. So it’s only logical that we work with whatever the community ends up creating around open source robotics, both to reinforce the community, and to hire from it.

That said, from a business model perspective it is more complicated. There are many companies that succeed with open source, but there is still no clear-cut model for building an open source company – and that tends to make people nervous. And yet by now we’ve seen several companies build on open source, including Google and Facebook, and it’s the same thing as building a startup on a LAMP stack: yes, you use open source and you may push back some code to open source, but you still have value you’re bringing on your own.

Earlier on we sold only hardware, which is easier to protect. Now that we’re getting into software sales, open source creates additional considerations for us, but it still hasn’t been an issue. Building on open source makes us more effective, and it certainly helps us engage much more with the robotics community.

Now that you’ve done all the legwork of developing the product and proving the market, do you have any concerns that another company with more resources could just come along and say, “We know who their buyers are, and we know what their markets are, and we can imitate their products …”?

Clearpath is sponsoring the 2015 Robot Launch “Open Source Robotics” Award, offering 10 hours of ROS consulting services to the winner.

ROS is open source and we contribute a lot back to it. We test a lot, we support people, we contribute drivers, and we contribute utilities back. However, the work that we do on autonomy and in controls –planning, mapping localization, interfacing, vehicle management – is not open source. We may open source parts of it in the future as we advance our state of the art, but the ‘secret sauce’ is still closed, even though it’s built on ROS middleware.

Could it make life easier for our competitors in some cases? Yes, but by the same token nobody makes money on having a better robot-logging framework, for example, and everybody saves money by having a good free robot-logging framework. If a better logging framework is your only competitive advantage, it’s not really an advantage, is it? Personally, I’d rather not spend money on a logging framework, and spend more to compete on the actual value that we bring to our clients.

You mentioned snow shoveling as one of your inspirations, which is a stereotypically Canadian problem. Did the Canadian climate and landscape figure into your business plan from the beginning? Is that why you wound up in field and service robotics, as opposed to some other application area?

Of course our robots can survive the winter – we’re Canadian!

One of our competitive advantages is the terrible weather we have in Canada. We have harsh winters, we have a lot of forestry, a lot of primary resources to mine and harvest, and we have a lot of space that is sparsely populated, which makes it easier to do testing. We also happen to have a very well developed technology ecosystem here. When you put those things together, it’s natural to lean towards field robotics.

When we started there were very few platforms out there that could actually do field robotics, and so we set that as our first niche.

How is the Canadian business landscape for robotics?

Many of our first customers were Canadian universities, some of which are leading the way in robotics integration and research. Meanwhile, Industry Canada is looking at how robotics technology can be adapted to existing manufacturing clusters in Southwestern Ontario, like the auto sector, which took a hit when the North American auto industry had a downturn. Canada also has strengths such as communications and optics, which naturally extend into robotics. It also helps when we can do business with Canadian suppliers; we get our GPS’ from NovAtel, which is based in Calgary, for example.

That said, as a Canadian company you’re forced to go multinational almost instantly because there isn’t enough of a robotics market in Canada to stay Canadian for long. We recently announced that we’ll be expanding office space into Silicon Valley and a primary reason for that is because of the opportunity that is available in the US market. Regardless, the Waterloo Region will continue to be Clearpath’s ‘home base’ for the foreseeable future.

People now refer to Waterloo as ‘Silicon Valley North’. What’s changed in the incubation space where Clearpath grew up since you launched?

It used to be that investors would demand that you move to the area where their office was, which was typically Boston or New York City or Silicon Valley, but that’s not always the case anymore.

It used to be that investors and venture capitalists would demand that you move to the area where their office was, which was typically Boston or New York City or Silicon Valley, but that’s not always the case anymore. Investors are starting to recognize that there are also benefits to being in other locations, especially for startups. Waterloo has a very dynamic tech community. It’s also a quarter of the price to buy a house here than in some of those larger US cities, and the time it takes me to get to work is 10 minutes, which makes it pretty livable and affordable compared to Silicon Valley. We found that VCs are now willing to accept that good ideas and businesses can grow and thrive in other areas as long as these areas have the right attitude – and Waterloo certainly has that.

Take the displacement of BlackBerry employees for instance – quite a few very smart people who were working at Blackberry have moved on to bring their skills and experience to other industries. We’ve hired a number of them, and a number of them have gone off to launch their own startups.

On top that is the continual increase in the brand of the Waterloo Region, and specifically the University of Waterloo (UW), as the place to go for getting a great education and practical experience that can lead to entrepreneurship. YCombinator’s Paul Graham has noted that UW has a fantastic record for turning out very strong companies, specifically hardware companies.

The other change worth mentioning is that Clearpath isn’t alone. The first group of startups that came out of UW includes Pebble, Vidyard, and more recently Thalmic Labs. These are all people who have graduated from UW’s engineering program within a few years of each other.

Does that help, too? Having a cluster that’s grown up together?

It certainly does.

Take recruiting, for example. When you’re hiring someone it helps prospective employees to know that there are other options. Clearpath has hired quite a few people from outside the region and if you’re taking a risk and moving a few thousand kilometers to join us, it’s good to know that there are other companies around, whether it’s for future growth opportunities, shared interests, or participating in knowledge transfer. For prospective employees, knowing that you can get back on your feet if you need to, or that your partner can find a job in the region, is very useful.

There is also something to be said for being part of a startup community. We all know each other, and we try to mentor each other. Even if you’re not working in exactly the same market, you can look at other startups and see them succeeding – and this is evidence that you can do it too.

Last week the Waterloo-based Clearpath publicly pledged not to develop lethal autonomous weapons, otherwise known as “killer robots”. In an open letter to the public, Clearpath’s CTO Ryan Gariepy wrote to support the Campaign to Stop Killer Robots, an international coalition that is spearheading the efforts to stop lethal autonomous weapons systems from being developed and deployed. While the Campaign has garnered significant support since its launch, it has not previously had support from the for-profit robotics sector – making Clearpath’s public statement a noteworthy demonstration of corporate responsibility , particularly given the company’s background in military applications. During a short interview with Gariepy at Clearpath, he summarized the letter as:

“When it comes to lethal weapons systems, not only do we not build them, we don’t believe anybody should build them. That’s our big statement here.”

But that’s just a small part of the story.

Clearpath’s statement is important because it demonstrates how the robotics industry can actively participate in the process of designing the ethical and legal landscape of our robotic future. It essentially sets a precedence of what corporate social responsibility will mean as more robots start to roll out.

Gariepy says,

We do feel that it’s the right thing to do. We’ve looked at potential risks to the business, and we feel that though there is risk to our business, this statement supports what our team believes … and what our partners believe.

Believing in something is one thing, but making it publicly known is another. And on the surface, many robotics companies will have more to lose than to gain from making a public statement on this issue, especially those (such as Clearpath) with military contracts. However, compared to large publicly traded companies such as Samsung (which manufactures sentry robots) or iRobot (which makes battlefield robots in addition to robot vacuum cleaners), Clearpath may be young and nimble enough to set such a precedent without long term negative consequences.

Only five years old, and not publicly traded, Clearpath made the decision to take a public stance on the Killer Robots issue based on the wishes of its 56 employees. According to Gariepy, the Canadian company has been following the issue closely since Christof Heyns (United Nations Special Rapporteur on extrajudicial, summary or arbitrary executions) published a report on this topic for the UN in April 2013. Right before Christmas of last year, Clearpath’s employees spent a Friday afternoon debating the topic of lethal autonomous weapons. This led to an internal poll to determine employee consensus on the issue, followed by the drafting of the public statement.

Clearpath is not saying that any and all military robots are bad, and they are very clear about this in the letter. In fact, developing robots for military application is at the heart of the company, which started off with a handful of University of Waterloo Mechatronics Engineering graduates building a robot to clear minefields. Mine-clearing robots are less controversial than lethal autonomous weapons systems, pointing to the existence of a grey area between the different uses of military robotics and whether they result in a net positive for the society. Yet Clearpath has drawn a line in that grey area – something no other robotics company has yet stepped up to do.

After the launch of the Campaign in 2013, the Campaign to Stop Killer Robots travelled to Canada urging the Canadian government to support the cause. According to Executive Director of Mines Action Canada Paul Hannon, the Canadian government’s Foreign Affairs and National Defence departments are still discussing the issue internally and have yet to formally declare Canada’s position. So I asked Hannon whether he thinks being a Canadian company would have made it easier for Clearpath:

Being the first commercial company in the world to make such a statement took a lot of courage I think, no matter where the company is based. It is clear from the statement that they have given it very serious thought and understand the impact it could have.

My personal view is that is probably harder for a Canadian company than an American one because so much R&D funding comes from outside of Canada particularly the US. Canadian companies likely need to develop partnerships with other non-Canadian firms to access those sources of funding. Taking such a principled stand could impact current and potential partnerships, so more credit to them.

 

My intuition is that Clearpath’s attention to roboethics issues will prove to be business savvy in the long run. Robotics has had a difficult time gaining social acceptance – largely because of the ways it has been portrayed in dystopian science fiction, but also because of the inhumanity it can project onto the battlefield. As more robots start rolling out into familiar places such as our homes and offices, the call for roboethics will grow stronger, and this will surely impact the kinds of robots we will allow our governments to deploy in military applications. Public trust is at stake, and the companies that are transparent about where they stand on roboethics issues stand a greater chance of social acceptance.

Roboethics will be a natural avenue for companies to show their social responsibility , and I remain optimistic that we will see more positive movement in this direction. Google’s establishment of an Ethics Board following their acquisition of the AI company DeepMind is perhaps further evidence of this.

I second Hannon who says,

Let’s hope that more companies from many countries follow their lead.

I have my fingers crossed.

A brief history of the Campaign to Stop Killer Robots

It’s been a long time coming in the history of military robotics to reach this point.

I first became interested in roboethics after learning about Samsung’s sentry robot used to monitor the demilitarized zone between the South and North Korea. That was almost a decade ago. Of course long before that, military use of unmanned aerial vehichles (UAVs) was on the rise.

With the persistent and increased use of UAVs in military applications, it became obvious that a natural next step in the technological evolution of the remotely operated UAVs was to develop highly autonomous versions of them. The discussion of lethal autonomous weapons systems among academics became heated, as you’ll find from the 2009 two-part robots podcast interviews with Noel Sharkey and Ron Arkin.

A public call to ban the development and deployment of autonomous lethal weapons happened more recently when the International Committee for Robot Arms Control (ICRAC) – one of the key player NGOs involved in the Campaign – drafted it in 2010. Back then, only a handful of academics and interested individuals signed up to support their cause.

ICRAC, along with Human Rights Watch, Article 36, and Mines Action Canada among other well-known international non-governmental organizations, started the campaign in April 2013, which got the United Nations to pay attention to the topic with the 117 states parties (87 countries attended the last meeting) of the Convention on Certain Conventional Weapons (CCW) this year.

Whether CCW will continue to work on the killer robots topic has yet to be decided, but the next annual meeting is scheduled for November of this year.

Small Unmanned Aerial Vehicles (UAVs) can be both safe and maneuverable, but their small size means they can’t carry much payload and their battery life only allows for short flights. To increase the range of a small UAV, one idea is to pair it with an unmanned ground vehicle (UGV) that can carry it to a site of operation and transport heavier cargo. Having both ground and aerial perspectives can also be useful during a mission. One challenge is to make sure the vehicles have the ability to rendezvous and perform coordinated landings autonomously. To this end, a paper by Daly et al. in Autonomous Robots presents a coordinated control method and experimental results for landing a quadrotor on a ground rover . The two robots communicate their positions, converge to a common docking location and the dock successfully, both indoors and out.

The video above demonstrates the use of a coordinated control strategy for autonomous docking of a Aeryon Scout UAV onto a skid-steer UGV (Unmanned Ground Vehicle) from Clearpath Robotics. The controller handles the nonlinearities inherent in the motions of the two vehicles, and is stable in the face of multi-second time delays, allowing unreliable wifi communication to be used in the landing. Both indoor and outdoor experiments demonstrate the validity of the approach, and also reveal the major disturbance caused by the ground effect when hovering over the ground vehicle.

For more information, you can read the paper Coordinated landing of a quadrotor on a skid-steered ground vehicle in the presence of time delays (John M. Daly, Yan Ma, Steven L. Waslander, Autonomous Robots – Springer US, July 2014) or ask questions below!

Robohub is an online platform that brings together leading communicators in robotics research, start-ups, business, and education from around the world. Learn more about us here. If you liked this article, you may also be interested in:

• HorseFly unmanned aerial parcel delivery system uses truck as base
• Amazon isn’t the only delivery service testing flying robots
• Amazon announces plans for drone delivery service, says safety will be key priority
• Quadrocopter failsafe algorithm: Recovery after propeller loss
• Is drone delivery really a robot startup?

See all the latest robotics news on Robohub, or sign up for our weekly newsletter.

by Ryan Gariepy

I recently spent over a week in Montreal talking with – and learning from – the Canadian field robotics community, first at the Computer and Robot Vision (CRV) conference and then at the NCFRN Field Trials. This was my first time attending CRV since I spoke at the 2012 conference in Toronto, and it’s quite inspiring how much robotics is spreading throughout Canada. Keynote speakers were Raff D’Andrea from ETH-Z and Frank Dellaert from Georgia Tech, both of whom had excellent talks. Bonus points to Frank for pointing out factor graphs to me for the first time, and Raff for bringing the best tabletop demo ever!

CRV transitioned right into the opening meetings of the NCFRN Field Trials. Or, for those of you who don’t like nested acronyms, the National Science and Engineering Research Council Canadian Field Robotics Network Field Trials. The NCFRN makes up Canada’s leading field robotics researchers, companies, and supporting government organizations, and has just marked its second year in existence. The first two days are filled with presentations, seminars, and poster sessions, and from there it’s into the eponymous field. Something which was both surprising and inspiring for me was seeing that Clearpath hardware represented every single ground vehicle and surface vessel in the poster session.

Spread from the Mars Emulation Terrain at the Canadian Space Agency to the east (aka, the “Mars Yard”) to the McGill MacDonald campus to the west, the field trials were quite the experience. Researchers and partners were able to see the algorithms operating live. They were able to collaborate on combining their technologies, and, in the most extreme cases, swapped code between robots to test how robust their approaches were. In my opinion, one of the most valuable aspects of the field trials is improving how testing and deploying autonomous vehicles in remote locations is completed. Teams’ preparations ranged from being as minimal and lightweight as possible to turning an entire cube van into a remote lab.

This years’ field trials also mark an important milestone for Clearpath Robotics: It’s the first time we’ve seen someone make coffee using our robots (at least, the first time we’re aware of it). Apparently it’s easier to plug a coffee maker into a Grizzly than it is to run an extension cord. I can understand why they were making coffee; this was the ASRL team which did 26 repeated traverses of nearly a kilometer each based on a single, visually taught route. And, since the ASRL has a thing for setting videos of our robots to music, here’s the research set to music.

I’ll close off by sharing the kind of experience which keeps us going: On the last day of the trials, I could only see six robots left rolling around the Mars Yard. There were three kinds, and all were robots designed and built by the Clearpath team – we’re thrilled to see that our work is having an impact on the robotics community and we can’t wait to see how Year 3 Trials will go next spring!