A closer look at Unitree’s growing line of quadruped robots

A closer look at Unitree’s growing line of quadruped robots

By Scott Simmie

 

If you follow InDro, you’ve likely heard bits and pieces about the Unitree line of robots by now. The Chinese firm specialises in quadruped robots – the ones that always remind people of dogs.

And while some of the Unitree line can indeed pull a few tricks, they’re serious machines. That’s why InDro became a North American distributor of the products. They’ve proven popular with clients, and InDro has done some serious modifications to enhance their capabilities for broader use-cases.

So we thought we’d take a spin through the Unitree line today, with some help from Account Executive Luke Corbeth. He knows these machines inside-out, and is usually the person behind the controls when we’re off at a trade show.

THE COMPANY

 

Back in 2013, there was no Unitree. But there was a student named Wang Xinxing, who worked hard on building a quadruped as part of his studies at Shanghai University. His vision? To build a powerful quadruped robot powered by low-cost, external brushless motors (think of the dog’s shoulders and hips). Wang took that vision and began working – and working.

He designed and tested legs. Worked on the robot’s control system, including designing motor drive boards, the entire master-slave architecture, the power supply – and much more. At the end of that long and focused process, a working quadruped Wang called XDog (where “X” means “mystery) began walking.

First, XDog was tethered in the lab. But before long, it was out in the wild.

Wang published some of that early R&D on a YouTube channel, which he still maintains. Here’s a look back to XDog in the lab. The description says this video was pulled together in 2014-2015.

PUSHING THE ENVELOPE

 

That student project – with a lot of further work – would eventually lead to the founding of Unitree in 2016. The firm became one of the first in the world to retail a quadruped. And it continues to put great emphasis on R&D. Here’s a snippet from its website:

“Unitree attaches great importance to independent research and development and technological innovation, fully self-researching key core robot components such as motors, reducers, controllers, LiDAR and high-performance perception and motion control algorithms, integrating the entire robotics industry chain, and reaching global technological leadership in the field of quadruped robots. At present, we have applied for more than 150 domestic patents and granted more than 100 patents, and we are a national high-tech certified enterprise.”

Now, Unitree sells multiple quadrupeds – with more on the way. 

Unitree Go1

GO1 EDU

 

This affordable robot comes with a lot of capabilites packed in.

Onboard EDGE computing is done by a Jetson Xavier NX. Five sets of fisheye binocular depth-sensing cameras allow the GO1 EDU to see its surroundings from the front, bottom, and sides. AI allows it to detect and classify humans. The robot can also walk alongside a person, rather than the “follow” mode often seen in similar machines.

The GO1 EDU is capable of navigating complex terrain, climbing stairs – even jumping over small obstacles. With a top speed of 14km/hour (3.7m/sec), the long-range GO1 can carry out even extended missions before requiring recharging.

And while the GO1 EDU is capable of carrying out inspection and surveillance work, there are other Unitree products more suited to that use-case. This machine, we’ve found, is best suited to those interested in R&D.

“The EDU version is designed to enable development,” explains Account Executive Luke Corbeth.

“So universities, corporate innovation centres, research institutes – anyone trying to find new ways for quadruped robots to understand their surroundings, plan their motion. Also ways to improve its gait, its ability to move in unpredictable terrain. Those are the sorts of things that are intriguing to this clientele.”

The GO1 has also proven to be appealing for proof-of-concept scenarios.

“Think of real-world applications like construction monitoring, inspections, security,” says Corbeth. “You can do that on a small scale on a budget, prove what you want the ideal workflow to look like, then scale up to one of the larger units.”

In addition to those capabilities, the GO1 EDU also has a playful side. There are a number of pre-programmed moves that make this robot suitable for entertainment applications. It has proven popular on stage with choreographed events and is (as we know) a hit at trade shows.

 

 COMING SOON

Unitree GO2

GO2

 

Unitree is set to release the GO2 shortly.

This machine does everything the GO1 EDU can do…and a whole lot more. If you look at the image above, you’ll notice the addition of a new sensor – right in the spot where a traditional dog’s mouth would be. That’s a LiDAR unit, and it significantly enhances the capability of the robot.

The L1 4D LiDAR sensor covers 360° x 90° in real-time. That means the GO2 can scan its surroundings in great detail, allowing for Simultaneous Localization And Mapping (SLAM) as the GO2 moves in its environment. With a mininum detection distance as low as 0.05 metres, no detail will escape the GO2. Equally impressive is the robot’s Generative Pre-trained Transformers (GPT), a neural network that helps GO2 understand and communicate with humans. (And if you’re wondering why they call that LiDAR “4D” – it’s because it can be tilted. So no, it’s not actually capturing in a fourth dimension – but it is versatile.)

The new machine, expected in Q4 2023, is also exceedingly nimble. There are some pretty serious algorithms onboard, and the quadruped is capable of descending stairs on its front two legs alone. Its 8,000mAh battery ensures it’s capable of long-range missions and a 15,000mAh ultra-long life battery is available as an option. The GO2’s voltage was bumped to 28.8V, which means this dog has more power and stability – and can trot along at an impressive 18 kilometres per hour. Software upgrades are carried out wirelessly from a cloud-based network.

“With the addition of the LiDAR unit, the GO2 will appeal to many in the R&D space – particularly those interested in SLAM and autonomous operations,” says Corbeth. “It’s also at a really impressive price point for its capabilities.”

He’s not kidding. Check this out:

THE BIG DOG

 

Looking for a serious robot for industrial applications? Something with a lot of power, and all-but impervious to even the most extreme weather conditions? Whether it’s remotely monitoring a key outdoor asset or making the rounds of a construction site or sensitive facility, the B1 is more than capable. The 50-kilogram machine is built for business, and can support payloads of up to 100 kilograms. It could easily carry critical supplies on a Search and Rescue mission, in addition to a myriad of sensors. (It can even carry a person on its back, as you’ll soon see.)

The machine is also exceedingly rugged. The B1 has earned an impressive IP68 Ingress Protection rating, meaning it can basically walk underwater – or brave a raging dust storm – with no issues. The robot has been built to withstand punishing conditions that would actually be dangerous for people (just one of the many advantages of robots). LiDAR is available as an option with the B1, allowing for SLAM, Search and Rescue applications, and more.

“This is the model that is most comparable to the Boston Dynamics ‘Spot’ – which is what most people think of when they think quadruped,” says Corbeth. “It also has a higher step height, which makes stairs a lot easier.”

Given its power and flexibility – multiple sensors can be added to the B1 depending on the use-case – this is the machine Corbeth feels is most suited to enterprise/industrial applications.

“This is the one I feel comfortable deploying into the real world at scale over a prolonged period of time. It’s a robust, dependable data collection asset that can be configured to excel at multiple applications.”

And when we said it could go underwater – we weren’t kidding:

 

SEARCH AND RESCUE

 

Even S&R applications – along with data acquisition to assist firefighters – are possible with a specially-outfitted B1:

WAIT – THERE’S MORE

 

The Unitree products are all great on their own. But InDro has developed an add-on that greatly enhances their capabilities. We call it the InDro Backpack, and it significantly improves the capabilities of the GO1 EDU, B1 – and will also be compatible with the GO2 at release.

In a nutshell, the backpack enables:

  • Remote teleoperation over 4G or 5G networks
  • Simple and intuitive interface for real-time, hands-on control of the robots
  • Monitoring of real-time individual sensor output in separate, configurable windows
  • Rapid integration of additional sensors without the hassle

We’ve written previously about the InDro Backpack, which is based on our highly popular InDro Commander.

“Out of the box, the Unitree GO1 has an app. But it’s not the greatest at managing all of the camera feeds,” says Corbeth. “Through the ROCOS dashboard, it’s a lot easier to see each of the feeds and get the most out of the impressive hardware that’s in the units. There are five sets of cameras and three sets of ultrasonic sensors – so we can really ensure the client is getting the most out of those.”

The software libraries onboard the Backpack also make any Unitree a fully ROS-enabled robot, which greatly expands their capabilities.

“That’s what makes Backpack valuable to the R&D community,” says Corbeth. “It means clients have access to all available packages to enable a wide range of applications, be it autonomous navigation, perception, motion planning, multi-robot systems – packages to ensure they can really jump-start their project. That’s the InDro value-add.”

 

InDro BackPack

NEED A HAND?

 

Or how ’bout an arm? Here, Unitree has you covered.

The company has developed the Z1, a highly dexterous manipulator. Lightweight but powerful, the Z1 has the option for multiple actuators/end effectors. Whether the task is opening a door or pick-and-place using optical recognition and AI, the Z1 can get it done.

“The world is built for humans and humans have the unique ability to open doors, move levers, press switches. A platform without a manipulator is incapable of interacting with the human world the way that a robot with a manipulator can,” explains Corbeth.

The Z1 can be mounted directly onto the Unitree B1. (We’ve mounted one and have to say it’s pretty impressive.)

“The Z1 is a highly capable manipulator with six degrees of freedom,” says Corbeth. “We are even looking at integrating it with some of the AgileX products we distribute.”

Plus, when compared with other robotic arms with similar capabilities, the Z1 comes in at an attractive price point.

“It’s exceptional value for money. I believe the Z1 will really reduce the barriers to entry – allowing clients to use this hardware to create proof-of-concepts, carry out studies, and just do general research and development with the unit.”

 

Below: Check out the Z1 in this Unitree video

INDRO’S TAKE

 

InDro is obviously pleased to be a North American distributor of Unitree products.

But we’re also pretty picky. We wanted to develop a relationship with a company that makes excellent products at a reasonable price, as well as a company that continues to push the envelope. Unitree was a perfect fit.

“If you look at their track record of new product launches and constant improvement and development, they work at a faster clip than anyone else. And that’s also a really good fit with how we do things at InDro,” says Corbeth.

“They’re constantly tweaking things so that their products are perfectly suited for the situations their clients want them to be used in. I’m personally a big fan of these robots – and our own customers have been really pleased, as well.”

You can find more details on the Unitree line – including downloadable spec sheets – right here.

And if you’d like a no-pressure conversation about how a Unitree might fit into your business or research plans, Luke Corbeth would be happy to chat and arrange a demo.

 

 

Smart Mobility companies invited to TCXpo 2023

Smart Mobility companies invited to TCXpo 2023

By Scott Simmie

 

We’re already excited. And we’ve blocked off September 27, 2023 in the InDro calendar.

Why’s that? It’s because that’s the day of the second TCXpo – a demo day and networking event featuring Canadian companies in the Smart Mobility space – will take place. The inaugural TCXpo, which took place at Area X.O in Ottawa, was amazing. And the second one will be even bigger and better.

The event is the only one of its kind in Canada. And it brings together cutting-edge companies with Connected and Autonomous Vehicles (ground robots, drones, cars) and related technologies. This year, more than 70 companies will demonstrate their products to the industry, investors, and government agencies in this space. 

Below: Some of the fun from the inaugural TCXpo in 2022:

Canada Robotics

A CANADIAN SHOWCASE

 

The goal of TCXpo is to demonstrate Smart Mobility technology. And Area X.O is the perfect venue. The R&D complex, operated by Invest Ottawa, is a private facility. It has its own Smart City infrastructure, including a dedicated 5G network. There are roads, traffic lights, railroad crossings – most of the things you’d find in an urban environment – except this site is dedicated for robots, drones and other Connected Autonomous Vehicles (CAVs). 

Now, picture a drone dropping a payload attached to a steerable parachute that can land within metres of the desired target. Envision an autonomous car, braking when a mock pedestrian appears before its path. Or a completely amphibious robot, capable of operating in water, on ice and snow, doing its thing for spectators.

All of these happened in 2022. And there will be even more at this year’s event.

“TCXpo is a tremendous opportunity for Canadian companies to showcase their products,” says InDro Robotics CEO Philip Reece (who was the person in charge of all aerial operations at the 2022 event). “This is such a forward-looking event. Many of the products shown here will play a role in the Smart City of the future.”

The event is put on by Invest Ottawa, Transport Canada, FedDev Ontario, Innovation, Science and Economic Development Canada (ISED), as well as other event sponsors (including InDro Robotics, one of the lead private contributors). 

And for Small and Medium-sized Enterprises (SMEs), it’s a major opportunity. Here’s how Area X.O describes what to expect:

“TCXpo will create an exclusive opportunity for invited guests to experience the power and impact of cutting-edge Canadian technologies, including many preparing for global markets. Companies large and small from Canada’s capital and across the country will showcase connected and autonomous vehicles, drones, smart-city solutions, IoT (Internet of Things), robotic technology, agri-tech innovations, and more. The showcased applications will span every sector of our economy – from intelligent transportation to defence, public safety, smart farming, telecommunications, cleantech, environment, and smart cities.”

Below: InDro Robotics CEO Philip Reece speaking at the 2022 event

 

InDro Robotics

HOW MUCH DOES IT COST?

 

Interested in getting your Smart Mobility product or service in front of the people and companies that count? Looking for investment? TCXpo is the place.

And the good news? Unlike expensive trade shows, those companies accepted can set up a display for free. (And while there’s truly no such thing as a free lunch, some really great food trucks showed up last year to feed the nearly 900 people who registered.)

 

GREAT! HOW DO I GET INVOLVED?

 

So glad you asked. First of all, have a good read about the event – including some metrics on attendance, media coverage and more – right here. If you think your company is a fit, head right over to the Area X.O intake form and let them know you’re interested. The deadline for applications is August 18.

It’s also worth noting that TCXpo takes place during Smart Mobility week in Ottawa. There are numerous other events, including the country’s premier Smart Mobility Conference – CAV Canada – taking place the same week.

Below: Check out highlights from TCXpo 2022

INDRO’S TAKE

 

We’re obviously big fans of TCXpo – and a key corporate sponsor. But there’s another reason we’re particularly excited about the 2023 event.

“The new Drone and Advanced Robot Training and Testing facility – which we call DARTT – officially opened in June,” says InDro CEO Philip Reece. “This will enable some truly great demonstrations of ground robots navigating challenging obstacles, as well as the ability to showcase new drone technologies within a safe netted enclosure.”

InDro – and all of our many robots and drones – will be at TCXpo 2023. We hope to see you there, as well!

DON DRONES ON ABOUT TRANSPORT CANADA CHARGES IN YOW INCIDENT

DON DRONES ON ABOUT TRANSPORT CANADA CHARGES IN YOW INCIDENT

By Scott Simmie

 

If you’re a regular reader (and we certainly hope you are), you may recall we recently broke the story about a plethora of fines levied by Transport Canada following two illegal drone flights at the Ottawa International Airport (YOW).

Those flights were detected by the YOW Drone Detection Pilot Program and Indro Robotics is the core technology provider for that platform. Both flights took place December 20 of 2022 and violated numerous sections of the Canadian Aviation Regulations (CARs).

The drone was a DJI Air 2S and police were directed to the pilot’s location by airport authorities. He was caught while actively flying and ordered to bring the drone down.

Both flights posed a risk. The first took place while a helicopter was landing; the second while a Jazz Q-400 passenger aircraft was coming in. Both flights – in addition to violating other sections of CARs – were well above the standard altitude limit of 400′ AGL.

Our story quickly gained attention in Canada’s RPAS world and piqued the interest of Don Joyce. He’s the person behind DonDronesOn, a YouTube channel with informative information for drone pilots.

Below: A look at the flight paths that were picked up by the YOW Drone Detection Pilot Program

 

YOW drone detection

CAUTIONARY TALE

 

Joyce sees this incident – as do many – as a cautionary tale. The potential for a conflict with crewed aircraft was very real. The drone was in the air as two different aircraft landed nearby.

“This is not an example of government overreach,” he says in a video you’ll see shortly. “Rather, a good example of technology and process applied to keep us safe from fools and bad actors.”

Joyce also rightly points out that drone detection systems are becoming more commonplace at airports and other sensitive facilities. Not knowing the rules is no excuse for those found caught breaking them.

“Drone detection systems are in use in Canada around sensitive locations like airports. They work. And they’re only going to get better and more widely deployed. This stuff is picking up both the electronic and acoustic signatures of our drones today.

“So if you think you’re flying with no one watching, think again.”

Joyce’s video triggered a lot of comments. Most were pleased to see the pilot was charged in this case. One commenter noted that – despite this incident – the number of rogue flights that blatantly violate CARs appears to have gone down over the years. (If you’ve been in this field for a while, you’ll recall crazy YouTube videos of blatant violations near airports, over crowds, etc.)

“I can say that in my experience over the past 10 years, there are less and less ‘idiots’ flying drones in Canada as many are indeed aware of TC regs and rules,” he wrote.

“And although all the TC regs/rules are not always followed, the ‘idiot’ flights placing other’s safety in question are extremely low today compared to five to ten years ago.”

Joyce has filed an Access to Information request to get the full file from Transport Canada. For now, he does a great job of explaining what happened – and how the pilot likely changed locations to evade DJI’s GeoFencing restrictions.

CONSEQUENCES

 

As we originally reported, the pilot was fined $3021 for violating seven sections of the CARs – including not having a Transport Canada RPAS Certificate. And that got us thinking: What other fines has Transport Canada levied recently in connection with violations of Part IX of CARs – the regulations governing Remotely Piloted Aircraft Systems?

Turns out, there have been a few. We take a look here at publicly available Transport Canada records for violations occurring in 2022. TC takes its time with these investigations; roughly a year transpired between most violations and the eventual fines.

Date of Violation: 2022/07/30 Location: Pacific Region

Though it took until June of 2023 for the offender to be served, the pilot was fined for violating three sections of CARs. According to Transport Canada: “A person operated a remotely piloted aircraft system (RPAS) when it was not registered and in Class F Special Use Restricted Airspace without authorization. A person also operated a remotely piloted aircraft system (RPAS) at a special aviation event or at an advertised event without a special flight operations certificate — RPAS.”

Given that this occurred in the Pacific Region, we believe this may have occurred at the Fort St. John International Air Show (which was underway at that time). The penalty assessed was $1400.

Date of Violation: 2022/05/29 Location: Quebec Region

Once again, three sections of CARs were violated. Says TC: “A person operated a remotely piloted aircraft system (RPAS) when it was not registered, and at a distance of less than 100 feet from another person measured horizontally. A person also operated an RPAS when that person was not the holder of a pilot certificate – advanced operations.”

The fine was served in May of 2023.

Date of Violation: 2022/06/17 Location: Quebec Region

Like the Pacific Region incident, this one also appears to have occurred at an airshow or special event. And there were a couple of interesting violations, including not having a Special Flight Operations Certificate and not having a copy of the RPAS owner’s manual available.

According to Transport Canada: “A person failed to operate a remotely piloted aircraft system (RPAS) in visual line-of-sight at all times during flight, and in controlled airspace. A person conducted the take-off or launch of a remotely piloted aircraft for which the manufacturer has provided a remotely piloted aircraft system operating manual without the manual immediately available to crew members at their duty stations. A person also operated an RPAS when that person was not the holder of a pilot certificate – advanced operations, and an RPAS having a maximum take-off weight of 250 g or more at a special aviation event or at an advertised event without a special flight operations certificate — RPAS.”

There were five CARs violations and a fine of $1500

Date of violation: 2022/06/19 Location: Quebec Region

This also took place at either an airshow or other special event – and these infractions set the pilot back by $900. “A person operated a remotely piloted aircraft system (RPAS) in controlled airspace,” states Transport Canada.

“A person also operated an RPAS when that person was not the holder of a pilot certificate – advanced operations, and an RPAS having a maximum take-off weight of 250g or more at a special aviation event or at an advertised event without a special flight operations certificate — RPAS.”

Date of violation: 2022/06/19 Location: Quebec Region

This incident involved violations of five sections of CARs. A fine of $1500 was levied in May of this year.

Here’s Transport Canada’s description of the offenses: “A person operated a remotely piloted aircraft system (RPAS) without a registration number clearly visible on the remotely piloted aircraft, in controlled airspace, and failed to operate it in visual line-of-sight at all times during flight. A person also operated an RPAS when that person was not the holder of a pilot certificate – advanced operations, and an RPAS having a maximum take-off weight of 250 g or more at a special aviation event or at an advertised event without a special flight operations certificate — RPAS.”

Below: The crumpled cowling of a Cessna. The aircraft collided with a drone operated by York Regional Police drone near Buttonville Airport on August 10, 2021. The pilot was later fined by TC. You can read our coverage of the incident here.

Cessna York Police Buttonville

Date of Violation: 2022/05/01 Location: Quebec Region

This incident involved seven infractions and a fine of $2100. Interestingly, this case involves the use of a First Person View device – where the pilot was wearing goggles and did not have a visual observer monitoring the drone directly (among other things).

Again, here’s the Transport Canada description: “A person operated a remotely piloted aircraft system (RPAS) when it was not registered, in controlled airspace, and at altitude greater than 400’ AGL without a special flight operations certificate — RPAS.

“A person also conducted the take-off or launch of a remotely piloted aircraft for which the manufacturer has provided a RPAS operating manual without the manual immediately available to crew members at their duty stations.

“A person operated a RPAS using a first-person view device without, at all times during flight, a visual observer performing the detect and avoid functions with respect to conflicting aircraft or other hazards beyond the field of view displayed on the device. A person also operated a RPAS at a special aviation event or at an advertised event without a special flight operations certificate — RPAS, and when that person was not the holder of a pilot certificate – advanced operations.”

Date of Violation: 2022/03/05 Location: Quebec Region

This one’s intriguing, as it involves an “unauthorized payload.” What that payload was is a bit of a mystery, as TC tell us that. However, this was part of a very pricy day: The eventual fine for violating five sections of CARs was $3950.

“A person operated a remotely piloted aircraft system (RPAS) when it was not registered, in Class F Special Use Restricted Airspace without authorization and failed to immediately cease operation when the safety of persons was endangered,” states the summary.

“A person also operated a RPAS while transporting an unauthorized payload and when the person was not the holder of a proper pilot certificate – small remotely piloted aircraft (VLOS).”

Date of Violation: 2022/03/16 Location: Quebec Region

Though details are scarce, we can read between the lines on this $1300 case and infer that someone flew their drone while First Responders or Law Enforcement were at an emergency scene. Costly mistake, along with not registering the drone.

“A person operated a remotely piloted aircraft system (RPAS) when it was not registered and over or within the security perimeter established by a public authority in response to an emergency. A person also operated a RPAS when the person was not the holder of a proper pilot certificate – small remotely piloted aircraft (VLOS).”

Date of Violation: 2022/03/16 Location: Quebec Region

Five CARs violations; $3780. Ka-ching.

“A person operated a remotely piloted aircraft system (RPAS) when it was not registered, in Class F Special Use Restricted Airspace without authorization and failed to operate it in visual line-of-sight at all times during flight. A person also operated a RPAS while transporting an unauthorized payload and when the person was not the holder of a proper pilot certificate – small remotely piloted aircraft (VLOS).”

 

OBSERVATIONS

 

As you perhaps noticed, the vast majority – eight of nine reported cases – occurred in Quebec. So one might immediately assume that pilots in that province are more reckless.

But we can’t say that from the data. Perhaps TC inspectors are more inclined to levy fines in that province, or there are more inspectors there. Maybe people are more inclined in Quebec to report drone violations to authorities. We really can’t say.

We did, however, find it interesting to note that some pilots were fined for violations such as not having a drone manual available on-site or wearing FPV goggles without a constant visual observer. It’s a good reminder that the regs are the regs – and they all need to be followed.

Below: Image shows the take-off points of the two flights detected by the YOW Drone Detection Pilot Project

YOW drone detection

INDRO’S TAKE

 

InDro Robotics was one of the first companies to offer hands-on drone training in Canada. We have trained police, firefighters, other First Responders – and more. We are also proud to have one of Canada’s leading online drone instructors, Kate Klassen, on staff.

Kate has trained more than 10,000 drone pilots in Canada. Her website, FLYY, offers everything to get pilots started – or take them to the next level for specialized training. (She is also a pilot and certified trainer for traditional crewed aircraft.)

“Regulations are there for a reason – to avoid conflict with crewed aircraft and to protect people and property on the ground,” says InDro CEO Philip Reece (who is also a private pilot).

“We’re pleased to have played a role in detecting these flights at YOW, and hope the fines levied do indeed send a message: Knowing and following the regulations is the right thing to do – and the best thing for this emerging industry.”

We should also mention that InDro is now offering basic and high-level drone training and evaluation in a massive netted enclosure at DARTT – the newly opened NIST-compliant facility for Drone and Advanced Robot Testing and Training at Area X.O in Ottawa. If you’re interested, you can contact us here.

You can find Transport Canada’s list of offences here. And we do recommend you check out Klassen’s FLYY.

Canadian Startup KiDrone has big plans – and technology – for reforestation

Canadian Startup KiDrone has big plans – and technology – for reforestation

By Scott Simmie

 

A Canadian startup has ambitious plans to deploy long-range, high-payload drones for reforestation at scale.

Using an extended-range, heavy-lift drone and patent-pending Machine Learning, the company calculates it could drop one million seeds in a single mission. Not only that, but it could plant seeds for different species in the most appropriate locations.

“We are a reforestation technology company,” explains CEO and Founder Trevor Grant. “We are going to be deploying heavy-lift unmanned helicopters coupled with AI machine learning to scale reforestation to industrial levels.”

That’s an impressive goal. Let’s look at how KiDrone plans to achieve it.

 

GENESIS

 

Many of the startups we’ve met over the years were founded by engineers. But CEO/Founder Trevor Grant is a lawyer by trade. So how did he wind up starting a venture involving drones and reforestation? Well, a couple of things happened.

First, he happened to watch a documentary on Netflix called Breaking Boundaries: The Science of Our Planet. It was about climate change, and the need to take urgent action on a global level. The following day, he happened to be reading an article about Beyond Visual Line of Sight flights. And then he started thinking.

“It tweaked in my head, perhaps the largest impediment to this (reforestation) is labour supply. And fleets of autonomous drones might be able to tackle such a problem at scale,” says Grant. Plus, he was also thinking about his children, living during an unprecedented era for planet earth.

“There was a genuine desire to leave the world to my kids better off than it was left to me,” he said during an interview at Toronto’s Collision conference.

And so KiDrone was soon born, with a mission to drop enough seeds to truly make a difference. But not just haphazardly. To ensure the best results, seeding would need to be targeted – with the correct species dropped in locations best suited to their survival and the broader ecosystem. Plus, the seeds would need to be coated.

“Seed encapsulation technology has been around for a very long time, but mainly in the agricultural sphere, not much for the reforestation or restorations sphere,” explains Fatima Mahmud, KiDrone’s Chief Scientific Officer. Mahmud is an environmental scientist born in the Middle East and who studied at the University of California, Berkeley, before obtaining her Masters degree in Toronto.

“Some of the reasons for encapsulating a seed, for aerial seeding specifically, is number one: It increases the flowability of the seed through the (dispersal) mechanism. Number two: It adds weight to the seed so the seed drops to the site. You can also add materials or compounds to the encapsulation that can deter pests and predators. And making the seed uniform allows it to find a suitable microsite in the soil once it’s dispursed.”

Below: Encapsulated seeds at KiDrone’s Collision display

 

 

 

KiDrone seeds

GETTING STARTED

 

Going from an idea to a viable product or service is a voyage – just ask any Startup. And the first part of KiDrone’s path has been to demonstrate that this is a viable, doable solution.

“It’s been a two-year journey to validate our hypothesis and validate where direct seedings works and where it doesn’t work,” explains CEO Grant. “Because direct seeding isn’t a cure-all for all reforestation needs. It’s highly effective in many situations – but not all.”

Post-wildfires (and Canada has had many this year), is a very promising use-case. Grant says high-intensity fires can consume the natural seed inventory that might be on the forest floor.

“So there’s a need for direct seeding at that point. Where direct seeding struggles is in drought-prone conditions,” he says.

 

TARGETED SEEDING USING AI

 

It’s not that difficult to deliver seeds via a drone. In fact, some companies have been dropping seeds and seed pods successfully. What differentiates KiDrone is its planned use of AI – and a proprietary seed dispenser capable of holding the seeds of 12 different species and disbursing them selectively. By examining multiple data points during flight, the drone will autonomously dispense the seeds best suited to particular locations based on the mission profile.

“There’s thousands of data points for any given site – climactic, GIS, various forms of imagery or LiDAR, soil lab results – an endless amount of data you can get to classify or gain conditions on a site,” explains Grant.

“And that will all lead to whether certain species may or may not be optimal (for a specific location), and what other species might be supportive. Our AI will be able to determine which trees are more likely to succeed in which areas. Because we’re not interested in monoculture or pine nurseries. We’re very interested in a more holistic reforestation approach that includes many different species, supported species, and Indigenous species of medicinal worth and spiritual worth.”

That last part is very important to KiDrone.

“Our biggest commitment is to work alongside the Indigenous communities where we operate. It’s their land and it’s their traditional territory. They should be the ones directing how reforestation happens. We simply view our roles as facilitating the reforestation goals that they have.”

Below: Founder/CEO Trevor Grant at the Collision conference.

 

KiDrone

THE NUTS AND BOLTS

 

It’s clear, speaking with CEO Grant, the company is working toward its goal via a methodical, evidence-driven trajectory. There’s been a lot of work on seed encapsulation so far though a partnership with the Northern Alberta Institute of Technology (NAIT), as well as a flight using a crewed helicopter for dropping encapsulated seeds. It successfully seeded 40 hectares.

“Our largest concern is validating where the seeds will grow, not where the drone will fly,” he says. “We’re concerned about the science behind encapsulation and determining where it’s effective to seed and where it’s not. We’re putting the science first, because we’re looking to do this for the next 20 years and not simply flip a carbon offset project.”

Existing startups doing seeding tend to use multi-rotor designs, which limits the distance they can cover per mission. Here, KiDrone differentiates itself by planning to use a much longer-range drone. The company has established a relationship with Scheibel, a manufacturer of UAVs (and landmine detectors). The company has a long-range uncrewed helicopter called the S-100. It can carry 50 kilograms of payload (seeds plus dispersal mechanism), and has an incredible range – up to 1000 kilometres, says Grant.

“The great thing about having such a large airframe that we’re looking to deploy – we’re able to carry 12 different species in one flight,” he says.

“So mid-flight, while travelling at 100 kilometres an hour, the system will be able to disperse an entirely different species, change the ratio of species being disbursed, add different nitrogen fixers or supportive species, all autonomously, based on AI and a seed-planting pattern that is pre-loaded to the mission.”

 

Below: A seedling that germinated from one of KiDrone’s encapsulated seeds. Image courtesy of KiDrone

KiDrone

THE BUSINESS CASE

 

There can be no question there’s an environmental demand for reforestation at scale. KiDrone’s pitch deck states “Reforestation in Canada is broken.

“Since 2010, Canadians have lost more than 44 million hectares of tree cover due to timber harvesting, wildfires, and commercial development. This represents an urgent, unmet need and opportunity to radically transform how industry & government deliver and scale reforestation in Canada.”

And with the devastating and deadly wildfires of 2023, the country has lost even more of that tree cover. The current system, of using human beings to plant seedlings, simply cannot keep up with the demand. It’s also inefficient – and there are vast tracts of forest in Canada that are simply inaccessible.

The big forestry companies also tend to have reforestation deficits, where they simply have not been able to reforest at a rate equal to the harvesting of timber. Plus, the KiDrone deck points out, “Corporate Canada’s demand for carbon offset opportunities vastly outweigh the current supply.”

So there’s not only an environmental imperative, but there’s also a strong business case. The company has been targeting three different sectors:

  • Top 12 Canadian forestry companies, each with reforestation requirements > 25k hectares
  • Federal & provincial forestry departments “focused on post-wildfire timber supply mitigation”
  • Carbon credit offset buyers and sellers

It all equals huge demand for a service like this, says Grant.

“Endless,” he says. “I think the wildfires we’ve had to date are a good example of of how large the reforestation required is in Canada alone – let alone globally.”

The company’s business model projects dropping 10k seeds per hectare in the future, with a 20 per cent viability rate. That comes out to 2,000 trees per hectare, at $.50 per tree. That’s $1,000 per hectare. Based on operating one drone and starting operations in 2024, its revenue projections climb to more than $1.2M by 2026 – and that’s with a single drone deployed. And because costs are low when compared with traditional tree seeding/planting methods, nearly all of that revenue would be profit.

The company is currently in a seed round (and we’re talking capital here, not trees), which its hoping to close late this summer or early fall. Once complete, there will be some additional immediate hires and KiDrone will be in “an early operational state.”

Grant is aware that BVLOS permission won’t be automatic, so he anticipates some of the early deployments will be VLOS, or operating with specific SFOCs.

INDRO’S TAKE

 

We’ve been through the Startup path, and know of the many challenges that come with the territory. But we also know a good idea when we see it. KiDrone has clearly identified its market and has laid out a solid path to commercialisation. It’s also a perfect application of autonomous technology for the Three Ds – taking on jobs that are dirty, dull and dangerous.

“In an era of climate change and with record-setting temperatures, getting more trees on the planet at scale helps all of us. I see this as definitely a Drones For Good application,” says Indro Robotics CEO Philip Reece. “I also really like seeing that KiDrone is taking it slow with an evidence-based approach – and a solid business plan. I look forward to hearing about their first deployment.”

You can learn more about KiDrone here.

Boston University uses AgileX LIMO for research

Boston University uses AgileX LIMO for research

By Scott Simmie

 

What will the Smart Cities of the future look like?

More specifically, how will the many anticipated devices operate – and cooperate – in this coming world? How will connected and autonomous vehicles interact to ensure the greatest efficiency with minimal risk? How might ground robots and drones fit into this scheme? And how can researchers even test algorithms without a fleet of connected vehicles, which would obviously incur great costs and require huge testing areas?

In the case of Boston University, the answer is in a small but powerful robot called LIMO.

LIMO

A VERSATILE PLATFORM

 

We spoke with three engineers from Boston University, each of whom are working with the AgileX LIMO platform. Before we get into an overview of their research, it’s worth taking a look at LIMO itself. Here’s how the manufacturer describes the product:

“LIMO is an innovative multi-modal, compact, and customizable mobile robot with Al modules and open-source ROS (Robot Operating System) packages, which enables education, researchers, enthusiasts to program and develop Al robots easier. The LIMO has four steering modes including Omni-directional steering, tracked steering, Ackerman and four-wheel differential, in line with strong perception sensors and Nvidia Jetson Nano, making it a better platform to develop more indoor and outdoor industrial applications while learning ROS.”

And it all comes in a pretty compact package:

 

LIMO

CAPABLE

 

While its user-friendly design is suitable for even enthusiasts and students to operate, its sophisticated capabilities mean it’s also perfect for high-end research. (You can find full specs on the product here.)

LIMO can detect objects in its surroundings and avoid them, and is even capable of Simultaneous Localisation and Mapping (SLAM). With a runtime of 40 minutes, extended missions are possible.

Here’s a look at LIMO in action, which provides a pretty good overview of its capabilities:

HIGH-LEVEL RESEARCH

 

We spoke with three people from Boston University, each of whom are using LIMO for different purposes. The three are:

  • Christos Cassandras, Distinguished Professor of Engineering, Head of the Division of Systems Engineering, and Professor of Electrical and Computer Engineering
  • Alyssa Pierson, Assistant Professor, Department of Mechanical Engineering
  • Mela Coffey, Graduate Research Assistant and PhD candidate under Alyssa Pierson in Mechanical Engineering

Cassandras is focused on groups of robots working cooperatively (and sometimes uncooperatively), called Multi-Agent Systems. If you think ahead to a connected Smart City of the future, the cars on the road would be Connected Automated Vehicles (CAVs). They would all be aware of each other and make autonomous decisions that ensure both safety and efficiency. Far enough down the road, today’s traffic signals, stop signs and more would likely not be needed because the vehicles are collectively part of a network.

“These vehicles become nodes in an Internet in which the vehicles talk to each other,” says Cassandras.

“They exchange information and so, ideally cooperatively, they can improve metrics of congestion, of energy, of pollution, of comfort, of safety – perhaps safety being predominant.”

In the video below, you’ll see LIMOs driving cooperatively, calculating in real-time the most efficient way to merge:

 

MAKING THE TRANSITION

 

But as we head toward this future, there will be a blend of regular cars and autonomous vehicles until the transition to automated driving is complete. And that period of transition will create its own challenges, which also interest Cassandras.

“So typically what we expect within the next, let’s say five to 10 years, is a mixture of the smart connected autonomous vehicles and the regular vehicles that we typically refer to as Human Driven Vehicles or HDVs. So the idea is: How can we get these teams of autonomous agents to work together?”

Obviously, testing this in a real-world scenario – with a blend of autonomous and HDVs – would be hugely expensive and require closed roads, etc. Enter LIMO – or, more accurately, a fleet of LIMOs.

“Since I can’t use dozens of real vehicles, I would like to use dozens of small robots that can be thought of as these autonomous vehicles, (which can) talk to each other, cooperate,” he says. “But also sometimes they don’t really cooperate if some of them are the HDVs. So what we are doing in our Boston University Robotics Lab, of which Alyssa and I are members along with several other colleagues, is we deploy these LIMOs that we have acquired as teams of autonomous vehicles.”

And what kinds of scenarios are they looking at? Well, consider how things work now. Cars stop at red lights, idle, and then quickly accelerate when the light turns green. This is hugely inefficient and adds to pollution. Wouldn’t it be better if there were no traffic lights at all, and vehicles could safely navigate around one another at peak efficiency? Well, of course. And that’s the kind of work Cassandras is conducting with a fleet of LIMOs at the Boston University Robotics Lab.

He is also one of the authors of a scientific paper that will be presented at the 7th IEEE Conference on Control Technology and Applications (CCTA) in August. That paper is entitled: “Optimal Control of Connected Automated Vehicles with Event-Triggered Control Barrier Functions: a Test Bed for Safe Optimal Merging.”

RASTIC

This fall, Boston University will open a new facility called the Robotics & Autonomous Systems Teaching & Innovation Center (RASTIC). There will be an area dedicated to mimicking a Smart City, with large numbers of LIMOs driving cooperatively (and sometimes uncooperatively). Cassandras says he intends to use a ceiling-based projector to create a simulated network of roads and obstacles on the floor for LIMOs to navigate.

“I envision about 20 to 30 LIMOs moving around, communicating – trying to get from Point A to Point B without hitting each other, as fast as possible, making turns, stopping at traffic lights if there are traffic lights, and so on… That’s the the grand vision. And RASTIC is intended for teaching as opposed to research.”

Other research using LIMOs will continue, meanwhile, at the existing Boston University Robotics Lab.

The following video, and this link, help explain RASTIC – and why this will be a significant facility for the Engineering Department.

THE IOT

 

So that’s one part of the resesarch using LIMOs. But wait, there’s more!

Assistant Professor Alyssa Pierson is also interested in Multi Agent Systems. But her work focusses less on the autonomous vehicle side of things, and more on general small-scale autonomous platforms. Think of delivery robots, drones, or even some other autonomous sensor platform making its way through the world.

“So thinking about instead of saying that two agents are inherently cooperative or non cooperative, what are all those nuanced interactions in between?” says Pierson.

“What does it mean if robots and a team have reputation that they can share among other robots? How does that change the underlying interactions? And we’re looking at these things, what reputation might mean, for instance, in perhaps robot delivery problems. How do they decide how to share resources, how to deliver supplies? The LIMOs provide a hardware platform to demonstrate these new algorithms that we propose.”

Graduate Research Assistant Mela Coffey is involved with this work, as well as some of her own as a PhD candidate.

Below: LIMO navigates obstacles, including dogs

LIMO Boston University

THE HUMAN FACTOR

 

Both Coffey and Pierson are also interested in how humans play a role in this world. And, more specifically, how robots might gather data that could assist humans in their own decision-making while tele-operating robots. Perhaps the robots might suggest that the human operator choose a more efficient route, for example.

It’s serious research, and a scientific paper on it has just been accepted for the upcoming International Conference on Intelligent Robots and Systems, IROS.

Coffey says the LIMO is perfect for this kind of research because it offers a hassle-free platform.

“From the start, they’ve been super easy to set up,” she says. “It’s nice just being able to take the robots right out of the box and there’s very minimal setup that we have to do. As roboticists, we don’t want to focus on the hardware – we want to just put our algorithms on the robot and show that our algorithms can work in real-time on these robots.”

Boston University has purchased a significant fleet of LIMOs from InDro.

“I think roughly total is about 30,” says Cassandras. “One of the things I was unhappy about with other small robots that we’ve worked with is that they would break a lot. That’s to be expected – if you have 10 and a couple break after a few months, that’s OK. But if you have 10 and six break, that’s not good. The LIMOs have been very reliable.”

 

ROS Robot

INDRO’S TAKE

 

Account Executive Luke Corbeth is the person who put these LIMOs into the hands of Boston University. He says it’s truly a perfect platform for such research.

Since LIMO is multi-modal, researchers can test their algorithms with a differential, Ackerman, omnidirectional or tracked system without needing to purchase 4 separate units,” he says. 

“The LIMO comes equipped with all the hardware needed for multi-robot teaming. It’s rare to find such a versatile and budget-friendly platform with the compute, connectivity, cameras and sensors that are needed to make this type of project possible.”

Corbeth deals with the majority of InDro clients – and is passionate about his work, and the work people like the Boston University team are doing.

I genuinely believe in a future where robots make our lives easier. My clients are the ones pushing us towards that future, so it’s satisfying to enable this sort of work. Above and beyond the research aspect, I know the students of my clients are learning a lot from using this robot, so it’s gratifying to know we’re assisting the next generation of innovators as well.”

And the best part? Priced at under $3000, LIMO is affordable, even for clients with limited budgets.

Interested? Learn more about LIMO and book a demo here.