InDro deploys drones, ground robots in Kelowna precision agriculture project

InDro deploys drones, ground robots in Kelowna precision agriculture project

Aug 1, 2024 | InDro News, Industry News, Photogrammetry, Scott Simmie

By Scott Simmie

There’s a buzz around Kelowna these days.

Well, actually, there are two kinds of buzz. The first is the occasional faint sound of a small but smart drone, carrying out flights every two weeks over two separate orchards. These orchards grow peaches, pears, cherries and more.

And the other buzz? Well, that’s the discussion this special two-year project – a collaboration between InDro Robotics and the City of Kelowna (enabled with funding from Agriculture and Agri-Food Canada’s Agriculture Clean Technology Program) – is generating among farmers in the region.

“Technology is always getting bigger and better,” says Riley Johnson, a manager at Byrnes Farms – one of the two locations where the project is being carried out. Johnson is an experienced farmer, and the land has been in the hands of his wife’s family for five generations. He knows the land and crops well, but is curious to see what additional data can be gathered.

“Anything in agriculture, the more information you get, you’re not going to be worse off. Especially for new farmers coming into the industry, any new kind of information can help out ten-fold – particularly if you’re on new land. Any information outside of the Old Farmer’s Almanac is always appreciated.”

What InDro is doing, to the best of our knowledge, is a type of precision agriculture that hasn’t been carried out before. It combines data acquisition from both drones and ground robots to ensure the most robust and reliable data possible.

This data is then used to assess overall vegetation health. Are there indications of pests in certain areas? Are any plants indicating low levels of chlorophyl? Does it look like that patch needs some pesticide – or maybe additional watering?

These are important questions for farmers whose livelihoods depend on maximizing the yield and health of their crops.

And this project? It’s all about finding the answers – and implementing solutions. Those solutions will include precision spraying of nutrients or other compounds in the precise location where they are required. The end result should be maximum yields with minimal – or no – areas of unhealthy crops.

Below: Healthy pears growing at Byrnes Farm. Photo courtesy Riley Johnson

Healthy pears in Kelowna. Photo courtesy Riley Johnson

AN INTRODUCTION

 

Before we get into all the details, it’s worth introducing you to Dr. Eric Saczuk (assuming you haven’t already met). He’s our Chief of Flight Operations – and comes with some serious chops.

Eric holds a PhD in Remote Sensing. He’s been assessing vegetation health (among many other things) using satellite data since the early days – long before drones came on the scene. But when drones did come on the scene, he quickly recognized their potential for the acquisition and interpretation of multiple kinds of data. He has flown missions all over the world on behalf of InDro, all involving complex data and analysis.

But that’s not all. He’s been an instructor at the British Columbia Institute of Technology since 2003 and has been the director of the institute’s RPAS Hub since 2016. He’s divided his time between BCIT and InDro since 2018 and is our go-to for highly complex operations. He’s also carried out multiple missions to acquire data for projects undertaken by Canada’s National Research Council – including this fascinating research on urban wind tunnels. There’s likely not a more qualified person in the country when it comes to drones and data.

Below: Dr. Eric Saczuk on an InDro mission. Image by Scott Simmie

SkyScout Ai Eric Saczuk

THE PROJECT

 

Okay. You’ve most likely heard of precision agriculture by now. When it comes to drones, most of us picture something like this: A drone with a multispectral camera flies over a field of wheat or some other crop. That multispectral camera captures spectrums of light both visible and invisible to the human eye. When that data is crunched, it provides a detailed picture of crop health (we’ll explore more of this shortly).

In collaboration with the City of Kelowna and local farmers, we’ve been flying a mission every two weeks over two separate fruit orchards. We use a drone with a special type of camera. It has five lenses. One of those lenses simply captures RGB (or simply, colour) images. But the other four have filters that are tuned to pick up light only within specific spectrums that can be collectively analyzed to indicate the health of vegetation.

“So in addition to the RGB camera, you’d have one camera capturing just red reflected light, one capturing just green reflected light using filters, and then the other two are what we call red edge and near-infrared,” says Dr. Saczuk.

Red edge is particularly useful in the early detection of disease or stress in plants – as it is highly reflected by healthy chlorophyl. But the real magic happens when you take the data captured in these different light wavelengths of light and run some calculations on them. That’s what gives you the bigger picture.

“Think of each of these images as a number. Capturing these multiple spectral bands allows you to combine them using complex equations in a type of calculator to give you various indicators of vegetation health,” he says.

That data can answer a lot of questions.

“Is it healthy? Is it not healthy? Is it being productive? Is there chlorophyll? If so, how active is it?” he says.

“These are the kinds of questions we can answer when we do what we call ‘multispectral band combinations.’ And it gives us a really clear picture that cannot be detected by the human eye.”

A CLOSER LOOK

 

We’re going to take a look at an image in a moment.

Before we get there, though, it’s worth exploring something Eric said earlier. He mentioned complex math. He’s right. The math happens at the individual pixel level between each of the input bands and there are millions of pixels per band. The most common calculation is used to produce a result called NDVI (not be be confused with NVIDIA, the producer of AI chipsets). The acronym stands for Normalized Difference Vegetation Index. And once the math is done, it gives you an accurate picture of vegetation health.

NDVI is calculated by using near infrared (NIR) and red bands (you’ll see the formula below).

The resulting pixel number in an NDVI image, will always be between -1 and +1. The higher the number, the better the health. If the number is low, it means there’s something worth looking at. The NDVI provides a detailed look at crop health and while it is regarded as the gold standard, there are also other calculations that can drill down to more specific indicators of vegetation health.

Below: Healthy vegetation absorbs most of the visible light that hits it, reflecting a large portion of the near-infrared spectrum. Unhealthy or sparse vegetation (right) reflects more visible light and less near-infrared light. When you do the math, it yields a lower NDVI number. (Public domain image by Robert Simmon.)

The second image is the equation used to calculate NDVI (which explains what those numbers are at the bottom of the first image).

NDVI
NDVI

FROM DATA TO DECISIONS

 

Using the equation above (as well as other formulas), Dr. Saczuk turns all of that data into something both meaningful and actionable. By looking at the data – and calculating not just NDVI but other indexes, images are generated that provide an at-a-glance look at crop health.

Traditionally, this has been a hugely time-consuming task involving multiple steps (and plenty of processing) on a laptop. Now, new tools are available that streamline the process. Dr. Saczuk is using a cloud-based solution specifically for precision agriculture.

“It really makes the whole process very efficient – because not only does it do the photogrammetry on the images, stitches them into these orthomosaics, but it also gives you the tools with which to analyse them. This would typically be a multi-step process, but this software makes it a one-stop shop, which is really nice.”

That’s without even getting into some of the AI capabilities of the software. It can, for example, count all the trees in a given orchard – and even give you the elevation of a specific tree.

Below: An NDVI image of one of the farms in the Kelowna project.

Solvi Kelowna NDVI

NOW WHAT?

 

At this phase of the project, InDro is gathering data by drone alone. But as it progresses, two more things will happen: We will introduce ground robots and precision spraying.

The plan is that a ground robot will initially be fitted with the same kind of multispectral sensor used by the drone. Autonomous missions will be plotted and the robot will capture a series of images from the ground as it drives through the orchard. That data will be crunched and compared with the results captured from the air.

“This is a way of doing ground-based validation of what we’re seeing from the air, from the aerial images of the drones,” says Dr. Saczuk.

Once that validation is complete and if problem areas are detected, the next phase would involve precision spraying – which could be carried out by an AGRAS agricultural drone – or even potentially by ground robot. Because all of the data is georeferenced, that means the fertilizer (or possibly pesticide or herbicide, depending on the issue) can be precisely applied to only the required locations. That, of course, is where the term Precision Agriculture comes from.

VICE-VERSA

 

This project is data-driven, with aerial and ground acquisition. But at the outset, shortly after our initial flight in April of 2024, farmer Riley Johnson noticed that a couple of trees weren’t doing well. It wasn’t clear what was causing this failure to thrive, but he didn’t want to take any chances that a potential disease might spread further in the orchard. So those trees were taken down.

In this case, because the issue was spotted early and the location was known, Dr. Saczuk is quite interested in doing some deep drilling into the data at that spot. In fact, that’s the very issue he has recently been exploring.

“So we’ve got that data, that information that’s saying, ‘Hey, these trees were actually not doing well.’ And then the next question is: Can we see anything in the multi-spectral images that would indicate that these trees are somehow spectrally or reflecting light differently than the ones that are healthy?”

This is something that is also of particular interest to Johnson. Will the data reflect what years of experience indicated was a problem to his naked eye?

“As the season progresses, it will be really interesting to see what InDro comes up with,” he says. “But I can definitely see the value of this for someone just getting into farming, or for farms up the hills with new plantings, new growth. This could be very useful.”

Below: Another image of a Kelowna orchard from this project, showing elevation

Solvi Kelowna Elevation

INDRO’S TAKE

 

We’ve been involved with precision agriculture projects in the past. In fact, we pioneered a “drone-in-a-box‘ solution, where we’ve shipped a drone to farmers. We talk them through the process of being a visual observer, then instruct them on how to power up the drone. InDro then carries out the flight remotely, using 4G or 5G – while in constant contact with the observer. When it’s done, the farmer puts the drone in the box and sends it back. InDro carries out the data analysis and quickly sends an easy-to-understand report indicating what areas require attention – and what kind of attention they require.

But this project is very different, and exciting for multiple reasons.

“The bi-weekly flights by drone will provide a huge amount of timely data, enabling us to detect any potential problems at an early stage,” says InDro Robotics Founder and CEO Philip Reece. “But by adding robots to validate from the ground, we’ll have a more robust dataset that can be used to truly pinpoint areas of concern and which may require precision spraying. We are going to learn a lot with this project – and believe our findings will be of great benefit to farmers down the road.”

A final note. When Dr. Saczuk isn’t carrying out these flights, they’re being flown by a new addition to the InDro team, Jon Chubb. He’s already had interest from other farmers in the Okanogan who are eager to maximize their own yields and have an early detection system for any trouble spots. If you’re in that neck of the woods and would like to arrange a demo, you can contact Jon here.

Sales and support: The InDro Experience

Sales and support: The InDro Experience

Jul 26, 2024 | InDro News, Industry News, Robotics, Scott Simmie

By Scott Simmie

 

We’ve all had them. Some unfamiliar number shows up on your phone, and the next thing you know someone is trying to pitch duct cleaning, a new cellular plan, or something else you likely don’t want or need.

You’ll never receive a call like that from us. For one thing, we don’t do duct cleaning (though we could probably build a robot capable of that). But there’s a bigger, more important reason. We cater – above all – to building a relationship with and satisfying the needs of the client. That’s both before we agree on a sale, and for long after the product is delivered.

While InDro is known for such ground-breaking stand-alone products as InDro Commander, the new InDro Cortex and InDro Controller, along with other platforms that enable researchers and R&D companies to build their own robots and drones with greater efficiency and simplicity, the bulk of our sales are custom products.

That means designing and building a robot or other device based on the specific needs and use-cases of the customer. Sometimes it’s a single, one-off design. But it could also be, as we recently reported, an entire fleet of custom robots for swarm research.

Regardless of the product or platform, the process nearly always begins with the first of many conversations with Luke Corbeth, our Head of R&D sales.

Below: Luke at a recent conference

Luke Corbeth

CORBY CAR CLEANERS

 

The story, or part of it, begins with a company called Corby Car Cleaners. “Corby” was Luke’s nickname in high school. An entrepreneur at heart, the prospect of working a traditional job didn’t appeal much. Corbeth wanted to learn about running his own business and was looking for the satisfaction that comes with building something from scratch. So he opened an on-demand car detailing business, where he’d show up at the client’s home and perform the service. How many people do you know that had the drive to start their own company in high school? (I can think of only one other; a friend who installed car stereos. I ran into him, decades later, at a CES show. His company was by then installing high-end systems – $100k+ – into cars owned by pro athletes, actors etc.)

“That was my way of building a basic business – doing sales and then actually executing the business,” he recalls.

With that – and high school – under his belt, Corbeth went to McGill University. Specifically, the prestigious Desautels Faculty of Management for a degree in Management. Corbeth had already shown a real knack for finance, which was part of the course. But he had the option of choosing specialisations. He was tremendously interested in technology – learning about AI, startups, the growing impact of other forms of tech – so he selected Digital Innovation as one of his tracks. But there was still another specialty he could select.

“I had something I was interested in (Digital Innovation), something I was good at (Finance) and something I felt I was missing,” he says. And that third piece of the puzzle?

“I didn’t understand the way operations are conducted. So I felt Operations Management was the missing piece; data science and understanding the optimal way to carry out operations. So that became the third piece of my arsenal.”

 

FROM CAR CLEANING TO WALL STREET

 

With his degree fresh in hand in 2019, Corbeth quickly landed two impressive jobs straight out of the gate. He did equity research at a hedge fund in New York, working to identify investment opportunities that the company referred to as having “material dislocation from fair value.” In other words, assets and stock that were undervalued. He also worked at a private real estate investment firm – which only had a single product to sell. He didn’t like that one much.

“When you have only one offering as a sales professional, it occasionally it feels like you’re trying to put a triangle into a circular hole – like you’re trying to force something upon someone.”

That wasn’t a fit for Corbeth. He was still drawn to technology; felt that was the future. Plus, something A McGill professor once said still resonated with him. It was along the lines of: ‘All the best investments in life, you need to be contrarian and right.’

“I was like, ‘Wow, that’s a very profound thought.’ Another way to think of that is: ‘You have to believe in a future that people don’t yet believe in, but will one day also believe in.’ And that’s what really attracted me to robotics,” he says.

“Sci-fi would suggest that it’s the future, but not everyone sees a future where robots are assistants to humans – carrying out useful and valuable tasks and doing the jobs we don’t want our kids to do. But I believe in that future, and I think people are going to want that future. And you know what? It’s becoming increasingly true.”

And then came the perfect fit: An opening for a sales professional at InDro Robotics in 2021.

“I was very excited at the prospect of joining the very field I was most interested in,” he says.

Below: Head of R&D Sales Luke Corbeth doing a public demo at the opening of Area X.O’s DARTT – the Drone and Advanced Robot Training and Testing facility 

 

 

THE INDRO EXPERIENCE

 

With this background, Corbeth had the perfect skill set – and passion – to take on sales with InDro Robotics. His keen interest in technology led him to quickly absorb everything possible when it came to sensors, compute capabilities, autonomy and more. But it was the chance to tailor something very specifically suited to the needs of a customer that truly appealed. He would never again be in a position of trying to put a triangle into a circular hole.

“What’s very unique about InDro Robotics is that because we take such a modular approach and we have so many partners and ways to bring solutions together, It feels like I always have the right shape to put in each hole – regardless of what shape it is.”

As mentioned, InDro most often builds completely custom robots for its clients. Corbeth is truly passionate about working with clients to fully understand their needs before even beginning to explore what options might be developed for them.

“The way I think about my job is this: People present a problem that they envision a solution for that requires uncrewed systems of some kind, whether that’s robots or drones. And I take it upon myself to thoroughly understand that problem and recommend all of the best available hardware and software at our disposal to solve that problem,” he says.

With rare exceptions, this isn’t just a simple phone call. Nor is it a process of simply ticking boxes for sensors, compute, platform etc. For Corbeth (and InDro), it’s the beginning of a relationship where together we explore every conceivable option to ensure that the client’s needs are met and their problem is solved. It generally starts with Corbeth sending a resource package to potential clients prior to any deeper conversation, so that they can fully understand InDro’s offerings, expertise, platforms, etc.

“The whole idea there is: Let’s identify the problem – and let’s identify the desired future. And since they’ve done their homework, ideally with the resources I’ve sent, then we can really dive into exactly what InDro Robotics can do specifically to get them there.”

This is a key part of the process.

“So the initial part of the conversation is always trying to understand the client’s desired future: What is it they’re actually trying to accomplish? And often times, just by talking it out with the client, I make discoveries about what they’re trying to do. They even make discoveries about what they’re trying to do. That’s always the first part: Thoroughly identifying the problem and what the ideal future should be.”

In some cases, for example, clients have come in looking for a wheeled platform, perhaps because they’ve seen a similar robot with wheels. But after further exploration of the use-case and terrain, they might together discover that a quadruped or tracked platform might be more suitable.

“If a client doesn’t know the existence of something, then it’s impossible for them to know that’s actually what they need,” he explains. “So part of my job is informing them of what is available, what is trending, what has been successful. It’s all about clarifying that problem and that vision for the future – and then seeing if we have something in the arsenal that can help get them there quickly.”

Below: Luke being interviewed at Area X.O’s DARTT

FROM PROBLEM TO SOLUTION

 

So then what happens?

Well, the second phase is seeing if there’s a fit. Can we provide what the client is seeking? Usually, of course, the answer is yes.

“With all that is available to us, can we offer something that meets their budget, meets their timeline, meets all of the project requirements so that we can help them achieve this future they’re seeking?”

This is something we really want to nail down.

“Sometimes in sales, companies will say: ‘Yes, my solution can solve every problem’ – and that’s not always the case. So if it makes sense to refer them to an adjacent providers outside our immediate circle, I’ll do that. But we’re in a fortunate position where we can solve a lot of problems – so that doesn’t happen very often.”

Once the client’s needs are fully understood, Corbeth works closely with them to drill down to the specifics: What type of LiDAR and other sensors best suit the application? How much compute power is needed? What platform and locomotion is best suited to the use-case? Is an autonomy stack required? The list, of course, goes on. And InDro has multiple options in every category to ensure the end product is the right one.

 

NEXT STEPS

 

We live, unfortunately, in a world where there’s a lot of hype – particularly when it comes to technology. There’s no shortage of vapourware, renders for products that don’t yet exist, and other techniques to draw clients into a sales funnel. At InDro, we take great pride in the fact our products are real and do exactly what we state.

“We’ve built custom products for everyone from small colleges and research teams right up to the largest technology companies in the world,” says Corbeth.

It’s one thing to say that; it’s quite another to prove it. And so Corbeth frequently arranges for an in-person or remote demonstration, where clients can control an existing robot or drone and take it for a test drive. They can experience the near-zero latency with remote ops, see the intuitiveness of the control system, and view all sensors, robot health (and much more) through the secure InDro Controller dashboard.

“We get them time on the sticks – ideally in person, but very often over Teams – and let them control a robot on our site. They can see how we plan missions, how we execute missions, and just how user-friendly and effective this technology is.”

That demo – controlling and seeing InDro products in action – generally seals the deal.

“Usually we move from there to quotes and service agreements and then, ultimately, a purchase order.”

Below: Luke engages with an attendee at the big IROS conference in Detroit in fall, 2023

FULFILLMENT AND SUPPORT

 

Even with the PO signed, the job isn’t done. Prior to starting the build, there’s generally a review call with the client along with engineering staff on the call. We want to ensure we have everything right, and understand every single facet of what the client is expecting.

“It’s a way of ensuring that what we end up delivering ultimately meets every expectation – because the worst thing you can do is send something out the door that doesn’t meet expectations. So we do that, and then we execute,” he says.

With the build complete, the product is shipped. Then there’s another call once it arrives. That ‘kickoff’ call points the client to the vast array of resources and continuously updated documentation InDro provides, as well as answering any initial questions the client might have. The whole concept here is to flatten the learning curve and ensure the client can get up and running with their robot or drone as quickly as possible. That support continues with any issues or questions that might arise after the client becomes familiar with their robot or drone or other device. At InDro, we are just as concerned with ensuring the client is satisfied and able to fully exploit their new product as we were with the original sale.

“We really value the client’s experience and the client’s feedback. In some ways, we create a feedback loop. And that feedback loop is what has enabled us to improve and evolve our products and our documentation at such a fast rate.”

 

KNOWLEDGE BASE

 

Luke Corbeth is frequently our lead person at robotics conferences and academic gatherings. He’s the guy PhD engineers will often engage with on InDro products and solutions. And the funny thing here? Luke’s comprehensive understanding of the technologies involved frequently lead people to assume he’s an engineer. In fact, that happened just this week at a University of Toronto event.

“That was like the penultimate compliment,” he says. “Despite having no formal education in engineering, I understand every part of our solution in enough detail that I can knowledgeably discuss hardware and solutions with some of the best professors in the world. Granted, I can’t build an autonomy algorithm. But I know what the robot needs to enable that.”

Of course, Luke would not have been able to reach that knowledge level without working closely with InDro’s talented engineering team at Area X.O. Whenever he’s had a question – they’ve had the answer. Many on our team have served as mentors for him.

Now, it’s one thing for us to say all this. But the real credibility comes with feedback from our clients. Luke received this note after the delivery of a custom robot build for, truly, one of the biggest companies in the world:

“Hey Luke – The robot is fantastic; the craftsmanship is superb; the power on the base is enabling; the intricate way in which the computer fits in the base housing is incredible; the compute box + mast feels “just right” (there’s no template for social robot design, but I feel like we got very close); all these things make me really confident that, with the right algorithms (my responsibility) we can safely and efficiently navigate through crowds.  It’s a really special robot that I can’t wait to put in the field.  Your team deserves a raise!”

(We’ll have to ask our CEO about that last bit.)

Below, our recent swarm robotics build for a US client

 

 

Robosense LiDAR Swarm

INDRO’S TAKE

 

InDro would not be thriving without satisfying our clients (many of whom are repeat customers, or are referred to us by existing ones). It is part of our very ethos to fully understand the problems they are trying to solve before we ever start discussing a solution. And it’s only then, through a very collaborative process with the client, that we can deliver a product that not only meets – but frequently exceeds – their expectations.

“Luke is a crucial and always enthusiastic part of the InDro Robotics team,” says Founder and CEO Philip Reece. “His understanding of the technology is phenomenal, and his drive to truly help clients achieve the future they envision seems to be hard-wired into him. Yes, meticulous builds are a key part of client satisfaction and InDro’s reputation. Luke plays an absolutely key role in ensuring that what we ultimately build and deliver is a match with their expectations.”

If you’re considering a robotics solution, contact Luke Corbeth here. He’ll be happy to arrange a test drive.

InDro completes design, build and delivery of robots for swarm research

InDro completes design, build and delivery of robots for swarm research

Jul 16, 2024 | InDro News, Industry News, Scott Simmie, Swarm Robotics

By Scott Simmie

You’ve likely seen swarm behaviour in nature – or at least video of it.

Flocks of starlings produce incredible patterns known as murmurations, undulating with split-second speed. Certain species of termites build massive and complex mounds, complete with sophisticated ventilation shafts to ensure air exchange and temperature regulation. Some types of fish, when attacked by predators, form tightly whirling balls as a defense mechanism. Bees build complex hives and have specific roles that all contribute to the collective. And ants, as you’re aware, can do a lot more than spoil picnics.

All of these, though very different, are examples of swarm behaviour. No single individual is in charge, yet these creatures work together to great synergic effect.

Below: A starling murmuration at Rigg by Walter Baxter, CC BY-SA 2.0 via Wikimedia Commons

A_starling_murmuration_at_Rigg_

SWARM ROBOTICS

 

It was actually through these observations in nature, particularly of insects, that the concept of swarm robotics first emerged.

“In a robot swarm, the collective behavior of the robots results from local interactions between the robots and between the robots and the environment in which they act… The research of swarm robotics is to study the physical body and the controlling behaviours of robots. It is inspired but not limited by the emergent behaviour observed in social insects, called swarm intelligence. Relatively simple individual rules can produce a large set of complex swarm behaviours. A key component is the communication between the members of the group that build a system of constant feedback. The swarm behaviour involves constant change of individuals in cooperation with others, as well as the behaviour of the whole group,” states this Wikipedia entry.

Much of the early research (and some of the more complex, current research) involves high numbers of very simple robotic devices. But as the field has developed, so too has the size and capabilities of the robots being used in swarm applications and research.

“The idea behind swarm robotics is the robots are able to communicate with each other and thus improve their decision-making and data acquisition in a given environment from a real world application standpoint,” explains Head of R&D Sales Luke Corbeth.

 

THE BOSTON EXAMPLE

 

One InDro client, Boston University, is using a very large fleet of small but smart robots called LIMO in swarm scenarios. One aspect of their research involves how autonomous vehicles interact with each other. Can the robots detect others merging and take appropriate action? How can autonomous vehicles detect and interact with robots that have been programmed to behave as if they’re human-driven vehicles?

Research like this is obviously more practical (and far less costly) to carry out in a lab rather than on city streets. It’s important work as we head toward the Smart Cities of the future, where eventually all vehicles will be autonomous – and much of the current infrastructure (traffic lights, stop signs) will no longer be necessary because all Connected Automated Vehicles will be part of a network.

“These vehicles become nodes in an Internet in which the vehicles talk to each other,” said Christos Cassandras, Distinguished Professor of Engineering, Head of the Division of Systems Engineering, and Professor of Electrical and Computer Engineering in an earlier interview with InDro.

“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.”

It’s fascinating work, and you can check out our story about it here.

But the ability of swarms to move and collect and share data collaboratively also opens up other use-case scenarios.

“If you are able to team your robots together, obviously you can cover a lot more area in a shorter period of time because the robots are essentially building a map together rather than being dependent on a single robot to build that map. One of the most obvious applications is search and rescue,” says Corbeth.

To illustrate, we borrow this example from research being carried out at Carnegie Mellon University:

INDRO’S CUSTOM SWARM

 

We recently built and shipped four identical custom robots for North Carolina State University’s Department of Mechanical and Aerospace Engineering. Corbeth was the conduit, putting together a package that suited their very specific needs. It began with selecting one of our 18 available platforms.

“They chose the Scout Mini, which is far and away our most cost-effective platform. And it’s unique in that it’s capable both indoors, like in their lab, but also for outdoor use,” says Corbeth.

That, of course, was just the start. Each robot was outfitted with the InDro Commander, which tucks a lot of power and flexibility within its IP54-rated enclosure. Commander houses the brains or the robot – which in this case is a very powerful piece of compute. Commander is completely customisable based on client requirements and contains ROS1 and ROS2 folders. (It also enables rapid integration and recognition of multiple sensors, and has the option of shipping with either or both of InDro’s autonomy stacks for outdoor or indoor use – think GPS-denied and SLAM. In this case, the client will be building their own autonomy stack, as that’s part of their research.)

And this client wanted Commander equipped with a powerhouse.

“They upgraded from a Jetson Xavier NX to the AGX Orin Developer Kit. That’s the best-in-class onboard compute you can add to a robot these days,” he adds.

That’s not hyperbole. The AGX contains a 2048-core NVIDIA Ampere architecture GPU with 64 Tensor Cores. Unfamiliar with Tensor Cores? They mean ultra fast AI processing.

“Tensor Cores enable mixed-precision computing, dynamically adapting calculations to accelerate throughput while preserving accuracy and providing enhanced security,” states the NVIDIA website. “The latest generation of Tensor Cores are faster than ever on a broad array of AI and high-performance computing (HPC) tasks. From 4X speedups in training trillion-parameter generative AI models to a 30X increase in inference performance, NVIDIA Tensor Cores accelerate all workloads…”

That AI computer is capable of – wait for it – 275 TOPS, or trillions of operations per second. (Yes, it’s kind of hard to get your head around that.)

 

SENSORS, SENSORS, SENSORS

 

Corbeth worked with the client to outfit each swarm robot with the sensors needed for the job. Starting with the basics, each robot contains two pinhole cameras (one front, one rear) so the operator can remotely see the robot’s eye view using our proprietary dashboard, InDro Controller. For depth perception, we added ZED X stereo cameras which – like human eyes – see the world in three dimensions. These are used for obstacle avoidance (in conjunction with LiDAR, which we’ll get to).

Connectivity is key with any robot. But when it comes to swarm robotics, it’s particularly important to have high-speed, low-latency communication not just with the controller, but between the robots themselves.

“Obviously doing swarm requires 5G connectivity,” says Corbeth. “So there’s a 5G modem in each robot as well as our Wi-Fi development module. In addition to communication between the robots and with InDro Controller, this also means they can program these robots without having to hard-wire into them. In fact, students or researchers could upload algorithms from pretty much any location – they don’t need to be on-site or on the same WiFi network, which has been the traditional route.”

Each robot features GPS and IMU modules, plus a Robosense Helios RS-5515 32-beam LiDAR module. These sense the environment in three dimensions by rapidly scattering and receiving the reflections from eye-safe laser beams. It allows the robot to safely navigate and map in completely unfamiliar and GPS-denied locations, and even carry out precision scans. We recently featured Robosense in this post, detailing how the quality is absolutely on par with the biggest names in the industry. (Selecting Robosense for this project saved the client close to $80,000 without losing any capabilities.)

Outstanding in their field: A look at the InDro-built swarm fleet just prior to shipping to the client

Robot Swarm

INDRO’S TAKE

 

It was a fascinating project to build a small swarm from the ground up for the specific needs of this client. We look forward to checking in with North Carolina State University once the research is fully underway – and will report back to you.

“There’s no doubt that swarm robotics is an important tool, not only in R&D but increasingly in real-world applications,” says InDro Robotics Founder and CEO Philip Reece. “Whether it’s search and rescue, mapping difficult environments at scale, or developing algorithms that will enable Connected Automated Vehicles in the Smart Cities of the future, swarms are synergic tools that produce outcomes far greater than the sum of their parts – and that do so with greater speed and efficiency.”

Interested in exploring options for swarm robotics? Contact Luke Corbeth here.

The new InDro Cortex: Amazing power in a tiny package

The new InDro Cortex: Amazing power in a tiny package

Jun 19, 2024 | InDro Autonomy, InDro Controller, InDro Cortex, InDro News, Scott Simmie

By Scott Simmie

 

The InDro Robotics engineering team, as always, has been hard at work. And we’re particularly pleased with our latest R&D breakthrough: The InDro command module, which we’ve named InDro Cortex.

It’s a small but exceedingly powerful box that will immediately enable robotic platforms to reach the next level. It’s robot-agnostic, meaning you can integrate Cortex with any ROS-based robot (which is pretty much all of them).

And what can it do?

A better question might be: What can’t it do? And the answer there is “not much.”

Below: An InDro Robotics Cortex mounted on a Unitree B2 quadruped. Cortex enables remote teleoperations, autonomous missions – and much more…

InDro Control Module ICM

BACKGROUND

 

There’s some important history here – which also tells you something about the InDro Robotics ethos.

In November of 2021, we released a very popular product. InDro Commander was developed in response to the needs of clients who wanted to be able to quickly turn a platform into a smart, easily customisable robot capable of teleoperation over 4G or 5G. From developers and academics through to large technology companies, our clients had been requesting such a solution. In fact, even in the development of our own products we identified the need for an all-in-one solution to speed robot integration and enable teleoperation.

“Our engineering team came up with the concept of the InDro Commander after integrating and customizing our own robots,” said Philip Reece, CEO of InDro Robotics, when the product was launched. “We realized there were hurdles common to all of them – so we designed and produced a solution. Commander vastly simplifies turning a platform into a fully functioning, teleoperated robot.”

Commander, as this webpage explains, “is designed to allow developers to quickly customize their own robotic solution. The following are common additions that can be easily added to any Commander-controlled robotic system: LiDAR (2D/3D), PTZ camera, thermal cameras, wireless charging, IMU, RTK.”

Commander, quite simply, takes the pain out of integration.

And how does it do that? Well, it ships with a powerful EDGE computer pre-loaded with ROS1 and ROS2 folders. It also contains multiple USB ports, allowing for the instant addition (and recognition of) pretty much any sensor you can throw at it. It allowed clients who were previously spending a ton of time on integration to simply attach Commander to the robot’s power supply, start plugging in sensors, and then remotely operate via a secure dashboard.

When it was first released, Engineering Lead Arron Griffiths dubbed it “ROS-in-a-Box.” We later changed that to Commander, but here’s the splashy video released at the time in advance of the big IROS show, where Commander made its debut:

SMARTER, FASTER, SMALLER

 

At InDro, we have a motto of “Invent, Enhance, Deploy.” But deployment doesn’t mean we rest on our laurels. InDro’s engineers continuously look at ways to enhance existing products. And that’s exactly what we’ve done with Commander. We asked ourselves: How could we make this even better?

“InDro Cortex grew out of the idea of making a general purpose backpack/module that any OEM could put on their motorized product – everything from a ground robot to even a golf cart, drone, car, truck or RC platform,” says Engineering Lead Arron Griffiths. “That single module makes any platform ROS-enabled, AI-ready and 5G connected.”

A lot of work went into this. One of the biggest tasks was designing and developing four new boards that pack a lot of AI compute power (and other features) into a tiny space. There are separate boards for EDGE computing, teleoperation data transfer, and sensor integration. There’s also a separate WiFi system.

“And of course, it can run on either a public or private 5G network,” says InDro Vice President Peter King.

And did we say size matters? The small size of Cortex means it can be placed on virtually any robot – from the smallest quadruped to the largest AMR. It can even be fitted to a drone.

“It’s tiny so it fits on everything. So basically we can turn any platform into a very powerful robot without all the time and hassle,” adds King.

Below: Much of the magic of InDro Cortex lies in the design, development and testing of four new circuit boards:

InDro Control Module ICM

AUTONOMY AND INDRO CONTROLLER

 

InDro Cortex is a powerful solution on its own. But pairing it with our new InDro Controller provides incredible synergy.

We just wrote a fairly comprehensive post about InDro Controller, so if you’d like more details you can check this out. But in a nutshell, InDro Controller is a powerful desktop UI for operating any robot. With a highly secure data connection, InDro Controller automatically senses all key parameters of any robotic device. It will automatically detect any sensors, providing a data feed from all of them with minimal latency.

“It can handle all sorts of data,” says Front-End Developer RJ Bundy. “It doesn’t matter if the robot has standard or custom sensors, InDro Controller can automatically detect and visualize them. For example, if you added a radar unit to an existing robot, the system will pick up on that immediately.”

InDro Controller provides the operator with a highly customisable dashboard for carrying out complex missions. It will also keep track of any missions manually carried out, and make them repeatable as autonomous missions.

And that autonomy? InDro has created two different autonomy software stacks. One is for outdoor missions that rely on GPS; the second is for indoor missions like SLAM in GPS-denied environments. That software comes with InDro Controller – and is also loaded onto the module. So the two key elements of this system – InDro Controller and InDro Cortex – have been made to operate seamlessly together.

“Anyone should be able to configure and control Cortex for their product, and indeed, their solution,” says Griffiths.

Below: A look at the powerful InDro Controller UI dashboard. The Controller software has been created to be an ideal match for InDro Cortex and is easy to operate.

Teleoperations with Robots

INDRO’S TAKE

 

We’re extraordinarily proud of the work our engineering team has put into Cortex – as well as InDro Controller. We believe these two, in combination, set a new bar for both robot integration and complex remote teleoperations.

“The design and creation of the boards used in Cortex was a significant technical challenge – and a significant achievement. The same can be said of our two new autonomy stacks,” says InDro Robotics CEO Philip Reece.

“Whether it’s a ground robot, a drone, or some other device that requires high processing power and minimal latency with all the features enabled by ROS, Cortex allows developers to quickly transform virtually any platform into a powerful autonomous robot with minimal effort.”

Interested in learning more? Of course you are. Contact our Head of R&D Sales, Luke Corbeth, right here.

 

 

New InDro Controller: A simple solution to complex robotics missions

New InDro Controller: A simple solution to complex robotics missions

Jun 12, 2024 | InDro Autonomy, InDro Controller, InDro News, Robotics, Scott Simmie

By Scott Simmie

 

InDro Robotics – as always – has been hard at work on innovative new products. And we’re particularly proud of our latest accomplishment: The InDro Controller.

It’s an all-in-one solution for operating virtually any type of robot from a highly secure console. It’s completely robot-agnostic, very easy to use – and exceedingly powerful. We’ll get into the details as we go, but first let’s hear from Front End Developer RJ Bundy with an elevator pitch.

“It’s an all-in-one data visualization, robot management and robot control software,” he says. “Whether you’re a student first learning how to use a robot or you’re a commercial giant, you’d be able to manage and maintain all of your robots.”

He’s not exaggerating. We walked through a demo of this system recently, with Head of R&D Sales Luke Corbeth at the controls and Bundy explaining the various features. We connected remotely with one of our InDro robots. The software immediately detected all of the sensors on the platform, offering up a display of windows showing the data they were collecting with minimal latency.

“It can handle all sorts of data,” says Bundy. “It doesn’t matter if the robot has standard or custom sensors, InDro Controller can automatically detect and visualize them. For example, if you added a radar unit to an existing robot, the system will pick up on that immediately.”

Of course, it has teleoperation. Missions can be run manually or autonomously (with InDro’s autonomy stack loaded onto any ROS-based robot).

“We have a GPS-based autonomy – which is better for outdoors – and then we have a SLAM- (Simultaneous Localisation and Mapping) based autonomy which is ideal for indoors,” says Corbeth.

And while InDro Controller has very complex capabilities, it’s a snap to use.

“Once it’s set up, it’s pretty straightforward to initiate the mission. Even someone without any robotics knowledge could do it,” he adds.

Below: The Pilot View mode in InDro Controller. Data from any desired sensors is displayed with minimal latency via a highly secure connection.

Teleoperations

HIGHLY SECURE

 

Regardless of whether you’re a startup, researcher, or a major corporation collecting sensitive data – security is important. InDro Controller has been built with that in mind.

“We’ve created an interface that makes it very hard for a third party to intercept any of those commands or the data coming from the robot to you.”

Though the person controlling the robot is the primary login, others with a secure login can also monitor the missions remotely from anywhere in the world. The software can store as many repeatable missions as you can throw at it, and you can initiate a previously stored mission with a single click.

 

MISSION PLANNING

 

Mission planning could not be simpler.

For the first mission, the pilot would manually control the robot. InDro Controller uses an Xbox controller plugged into your computer for intuitive operation (though other options are available). All buttons on the Xbox device can be quickly mapped to carry out specific functions.

InDro Controller tracks everything you’ve done – and we mean everything – and saves it as a repeatable autonomous mission.

“If you manually drive the robot somewhere, it will remember it’s been there and it’ll be able to go back, follow that same path every time. It will also remember to carry out any specific actions you’ve taken at those points of interest, including camera angles, zoom, etc.” says Bundy.

The mission planner also automatically loads a map to locate your robot (you also have the option of satellite view), so you can monitor exactly where it is on any given mission.

And, of course, it can do this for an entire fleet of robots.

Below: InDro Controller shows Points of Interest – which can be repeated with saved missions

Robot Teleoperation

HIGHLY CUSTOMISABLE

 

InDro Controller has been designed to allow users to easily customise the user interface for any robot, any mission, and any dashboard view. Multiple streams of data, including upload and download speeds, battery levels and overall robot health are available at a glance. Oh, and did we mention it also works with third-party autonomy stacks?

“The dashboard, the cameras, the heads up display on the autonomous missions – those all can be customised,” says Bundy. “We’re also adding other personal user customisations, like a light and dark mode, metric conversion, schedule missions – all the kinds of features you could want.”

InDro Controller already works exceptionally well. But – as with all of our R&D projects – it will continually be refined with additional features and functionality.

“We’re heavily invested in continuously improving the software,” says Corbeth. “So regardless of which version you’re shipped, know that this is something that InDro Robotics is constantly developing and improving with client feedback in mind to ultimately provide the best mission planning, teleoperations and development software tool in the robotics industry.”

Speaking of versions, there’s a simpler version of InDro Controller – which does not have the autonomy features – already being used for missions in the academic world. Feedback has been excellent.

“Users tell us they find it InDro Controller Lite exceedingly powerful, but also very simple to use. That was exactly our goal in developing this product,” says Corbeth.

Academics and corporate innovation groups could take advantage of the Lite version, while the InDustrial package is intended for solving more complex problems in an industrial environment

 

SATISFYING R&D

 

For Front End Developer Bundy, who oversaw this project with support from other engineering staff, it’s been a hugely satisfying – and challenging – project.

“A lot of the customization features were pretty difficult because it has to be robust and dynamic, which is always tough,” he says. “This is a relatively complicated application and I’ve managed to put together something pretty nice and functional – and it will only get better. “I’ve had a bunch of other help, but putting together the UI for InDro Controller has been, and continues to be, highly satisfying.”

Teleoperating Robots

INDRO’S TAKE

 

We’re obviously excited about InDro Controller. And we’re particularly excited because we have a forthcoming piece of hardware – the InDro Module. It’s a small box with a lot of brains that can be added to any robot to increase functionality and enable the seamless addition of sensors and other modifications (as well as pre-loaded autonomy stack and ROS drivers). We’ll have more on that soon, but it’s the perfect match for InDro Controller for users with complex requirements.

For the moment, we’re looking forward to putting both the Lite and InDustrial versions into the hands of clients.

“When we first began remote teleoperation several years ago, we relied on third-party software as the UI,” says InDro Robotics CEO Philip Reece. “But we found it wasn’t powerful or customisable enough for our needs. It also required that we have our own autonomy stack – and we did – but many clients do not. InDro Controller comes with our proprietary autonomy stacks for both outdoor and GPS-denied locations. And, as noted previously, this is a long-term project, where even early adopters can be assured the package will be continuously refined with additional features.”

Word has already been spreading in the R&D and commercial fields about this product, and the feedback from those using the Lite version has been outstanding. Interested in learning more or seeing a demo? Contact Luke Corbeth here.

InDro Robotics ROS-based drone an R&D powerhouse

InDro Robotics ROS-based drone an R&D powerhouse

Apr 16, 2024 | InDro Forge, InDro News, Innovation, Scott Simmie

By Scott Simmie

 

InDro Robotics is pleased to unveil details of its highly capable new R&D drone.

Running the Robot Operating System (ROS) and with powerful onboard compute capabilities, the drone is perfect for advanced Research and Development.

“It’s a drone geared toward R&D first and foremost,” explains Luke Corbeth, Head of R&D Sales. “It truly is a flying robot – and you can program and use it in a very similar fashion to all our other robots.”

There’s a real demand in the research world for open-source drones that can be programmed and run highly complex algorithms. These kinds of drones can be used to study swarm behaviour, object detection and identification, mapping in GPS-denied locations and much more.

For some researchers, the budget go-to has been the Crazyflie, a micro-drone that uses a Raspberry Pi for compute. Its advantage is that it’s quite affordable. But its low cost, 27 gram weight and relatively low computing power means it has limitations – including the ability add sensors of any weight.

“This drone can do so much more,” says Corbeth. “With the NVIDIA Xavier NX onboard for compute, it can effectively map entire environments. And when it comes to landing and object recognition, it’s truly phenomenal. It can even land on a moving vehicle.”

Below: A look at InDro’s new drone, which comes complete with LiDAR, a depth-perception camera, 5G connectivity – and much more.

InDro ROS drone

THE BACK STORY

 

If you’ve been following the latest news from InDro, you’ll be aware we have an incubation agreement with Cypher Robotics. That company builds solutions for cycle counting and precision scanning in the industrial/supply chain space. InDro assisted with the development of its signature product, Captis.

Captis integrates an autonomous ground robot with a tethered drone. As the Captis robot autonomously navigates even narrow stock aisles, the drone ascends from a tether attached to that ground robot. The drone then scans the barcodes (it’s code-agnostic) of the products on the shelves. All of that data is transferred seamlessly, in real-time, to the client’s Warehouse Management System (WMS), WCS (Warehouse Control System) and WES (Warehouse Execution System) software.

The capabilities of Captis led to a partnership with global AI fulfilment experts GreyOrange and leading global telco innovator Ericsson. The product debuted at the recent MODEX2024 conference (one of the biggies in the automated supply chain world), where it gained a *lot* of attention.

While working on the project, it was always clear the drone – thanks to multiple modifications – would be highly suitable as a research and development tool. It’s capable of machine vision/object recognition, machine learning, and can find its way around in completely unfamiliar, GPS-denied environments.

“In fact, I have one client that’s using it for research in mines,” says Corbeth.

 

THE JETSON DIFFERENCE

 

NVIDIA has made quite a name for itself – and quite a profit for its shareholders – with its powerful AI-capable processors. The Jetson Xavier NX features a 6-core NVIDIA Carmel Arm®v8.2 64-bit processor running at speeds of up to 1.9 GHz. Its graphics processor unit features a 384-core NVIDIA Volta™ architecture with 48 Tensor Cores. Put it all together, and the computing power is astonishing: The Xavier NX is rated with a maximum achievable output of 21 TOPS – trillion operations per second. (We were going to try to count, but thought it more efficient to rely on NVIDIA’s specs for this.)

The LiDAR unit currently shipping with the drone also has some flex. It’s the Ouster 32-channel OS1 (Rev6.2). With a maximum range of 200 metres (90 metres on a dark, 10 per cent target), its powerful L3 chip is capable of processing scans of up to 5.2 million points per second with 128 channels of vertical resolution (again, we didn’t count). Hostile environment? No problem. The LiDAR can operate from -40°C to 60°C and has an IP68 Ingress Protection rating.

“The OS1 is designed for all-weather environments and use in industrial automation, autonomous vehicles, mapping, smart infrastructure, and robotics,” states its manufacturer. “The OS1 offers clean, dense data across its entire field of view for accurate perception and crisp detail in industrial, automotive, robotics, and mapping applications.”

The unit uses open source ROS and C++ drivers, and comes with Ouster’s Software Development Kit. Its ability to accurately sense its environment (down to distances of 0.5 metres away), combined with the NVIDIA processor and the depth camera also allows this machine to do something pretty extraordinary: It can recognise and land on a moving platform.

“That’s a very challenging problem to solve and requires not only specific sensing but also really powerful onboard compute. This drone can do it,” explains Corbeth.

Already, word about the product has been spreading. A number of units have already been sold to academic institutions for research purposes – and the team has been hard at work building and testing for the next set of orders (as seen below).

THE FORGE CONNECTION

 

Like all new products, the new drone required custom parts. We looked no further than InDro Forge, our rapid prototyping and limited production run facility in Ottawa.

Using state of the art additive and subtractive tools, the Forge team created custom mounts using carbon fibre and other strong but lightweight materials, while also ensuring the frame was robust enough to take on even the most challenging environments where these drones will be deployed.

“InDro Forge has been critical to the finished product,” says Corbeth. “We wanted a look, feel and quality that matches this drone’s capabilities – and InDro Forge delivered.”

InDro ROS drone

INDRO’S TAKE

 

We’re obviously excited about the capabilities of this new drone, and we’re not alone. Interest in this product from researchers has already been significant. In fact, we’re not aware of any other drone on the market offering this combination of specific capabilities.

It was that void – in concert with our partnership with Cypher Robotics – that led to its creation.

“InDro has always placed a great emphasis on the development of innovative new products,” says CEO Philip Reece. “We build new products at the request of clients and also develop our own when we see a market opportunity. In this case, the requirements for Cypher Robotics dovetailed nicely with demand for such a drone from researchers.”

Production of the new drone is moving at a swift pace. If you’re interested in a briefing or demo, you can contact us here.