By Scott Simmie

We’ve all seen the videos.

Quadrupeds doing parkour. Humanoid robots dancing, throwing punches, or effortlessly picking up boxes.

It’s easy to assume those capabilities come straight out of the box. In reality, almost every successful deployment involves configuration, software development, sensor integration, testing and refinement.

So while all that kicking and dancing may look impressive, it leaves out a very critical part of the equation. Before a robot can do any of those things reliably, it has to be built, programmed, trained and – most importantly – tested in the environment where it’s actually going to work.

That’s the part most of those videos never show: Mud that behaves like glue. Loose gravel that jams up tracked platforms. Dust, rain, extreme temperatures and uneven terrain. Those are the things that determine whether a robot succeeds or fails.

At InDro Robotics, we build for the real world – every step of the way.

Above: Our new Prowler wheeled quadruped inspection robot gets put to the test at Ottawa’s Area X.O. Below: An earlier iteration of our Sentinel is tested in water of varying depths

USE-CASE

At InDro, we don’t simply sell robots. We customise them for each and every customer.

During the initial “discovery” call, our Head of R&D Sales Luke Corbeth learns as much as possible about the client’s needs. Is the robot for indoor use only? Remote asset inspection (where it will undoubtedly be exposed to inclement weather)? Does it need to climb stairs, navigate rough terrain, identify objects?

That first conversation isn’t about choosing a robot; it’s about understanding the problem. Once Corbeth has a good sense of use-case, he can start recommending platforms and parts suitable for the job. If the robot is going to be outdoors, for example, he’ll suggest platforms and components that have a solid Ingress Protection (IP) rating.

The general baseline for outdoor robots is IP65, meaning they’re capable of handling light rain, irrigation sprays, dust etc. But a more demanding environment could easily call for a higher IP rating to ensure it can handle mud, temporary submersion etc. Now we’re into IP67+ territory.

Sensors are integral to every deployment. Depending on the application, a robot might carry anything from a simple RGB camera to LiDAR, depth cameras, thermal imaging, GNSS/INS modules or PTZ systems. These feed data into InDro Cortex, where onboard AI fuses the information, giving the client the complete picture via our InDro Controller user interface.

TESTING, TESTING, 1,2,3

Simulations are an important first step. They let us recreate weather conditions, terrain and mission scenarios. But simulations only go so far. Eventually, every robot has to prove itself in the real world.

We ensure they’re ready by putting them through their paces at Area X.O’s Drone and Advanced Robotics Training and Testing zone, or DARTT. It’s a series of testbeds that include sand, gravel, uneven terrain, water, an adjustable incline ramp and more. All of these have been built to the demanding requirements of the US-based National Institute of Standards and Training (NIST). To the best of our knowledge, it’s the only such facility in Canada.

But we don’t stop there. If the client has particularly demanding terrain or surfaces, we test the robot at their location when possible. And in those cases where we can’t, if there’s an issue we will troubleshoot remotely as a first step. In the rare event where a robot is not operating as planned in the field, we will either address the issue back at our R&D headquarters or dispatch a field engineer to the site.

And yes, despite all that, real-world testing occasionally produces unexpected results. In one deployment, deep mud inside a mine proved too much for a quadruped platform. It simply became stuck. Rather than treating that as failure, we treated it as engineering data – returning to the drawing board to improve the solution.

Below: Testing Sentinel onsite in real-world conditions in Ottawa

INDRO’S TAKE

We take building robots very seriously – and our top priority is ensuring that the robots we build meet the needs of our clients.

“Reliability isn’t something you add at the end,” says Founder and CEO Philip Reece. “It’s built into every stage of development – from that initial discovery call through platform selection and testing through to deployment. That’s one reason why so many of our customers come back when they’re ready for their next robot.”

Anyone can demonstrate a robot for five minutes. The real challenge is making sure it performs every day, in conditions that are rarely ideal. To set up a no-pressure discovery call, contact Luke Corbeth here.

At InDro Robotics, we specialize in cutting-edge autonomous systems, including our flagship Sentinel inspection robot. We needed a vision-based perception system that could deliver high-performance 3D mapping (SLAM) consistently across diverse and challenging environments—from indoor facilities to high-speed outdoor operations—all while maintaining an efficient budget.

We found our solution in the Stereolabs ZED X camera system. After four years of leveraging Stereolabs technology, the ZED X has become a core component of our advanced autonomous fleet.

The Impact on Our Robotics

The integration of the ZED X directly addressed our core engineering challenges and delivered significant, measurable benefits:

• Exceptional Performance-to-Cost: The ZED X offers the ideal balance of powerful visual SLAM capabilities and cost-efficiency, allowing us to deploy advanced technology without compromising our budget.
• Enhanced Outdoor Capabilities: The integrated polarized lens was essential for our outdoor Sentinel Security Robot, drastically improving visual clarity and reliability in complex lighting conditions.
• Faster Deployment Times: Measurable improvements in mapping speed and accuracy directly lead to reduced deployment times for new robotic systems, accelerating our project turnaround.
• Accelerated Development Cycles: Robust ROS 2 SDK support and simplified integration enabled our engineering team to focus on innovation and advanced autonomous behaviors, rather than vision system troubleshooting.

As our Director of Engineering, Arron Griffiths, notes, the ZED X represents “a significant leap forward in performance and integration,” proving the value of strategic technology partnerships in driving innovation.

Read the full story HERE to learn how our partnership with Stereolabs enhanced the functionality and reliability of our autonomous robot fleet.

By Scott Simmie

So. You want to buy a robot. 

Maybe you’re in the academia or R&D fields, and you need a robot for research. Or you’re a commercial/industrial client, looking for a device to carry out regular maintenance inspection, surveillance, or some other repeatable data acquisition task. You might even be looking for a robot that can pick things up and move them from one location to another. 

Regardless of the use-case, you’ll have a budget. And you’ll want to make the most of it. 

The question is: How, precisely, do you do that? How can you ensure you’re getting the best bang – and robot – for the buck?

At InDro, we take pride in how we work with clients to ensure they’re maximising their spend while we minimise their pain. Regardless of where you ultimately purchase a robot, we hope the tips this post reveals will help inform your choice.

Below: ‘Rosie’ – a highly complex dual-manipulator robot InDro built for a client

BUY ONLY WHAT YOU NEED

It’s a big world out there. And there’s more than one place to go robot shopping. But many suppliers lock clients into “packages” – which include a fixed number of sensors. Those packages tend to be rigid, even if the client might already have some of the desired components.

“They tend to bundle in components, be it expensive cameras or sensors or compute that the client already has,” explains Head of R&D Sales Luke Corbeth.

At InDro, he says, there’s an important difference.

“What we’re able to uniquely do is custom outfit a robot to only include what they need – so they don’t need to repurchase a 3D LiDAR or, say, a depth camera. If someone says, ‘Oh, I want this package but don’t need an IMU because I already have one,’ then we can bump down the cost and just not include the IMU.”

BUILD MODULAR

Needs evolve over time – as does technology. So it’s quite likely, during the lifespan of your robot, that more powerful sensors will be released. Or perhaps your own needs will change, and you realise that adding a thermal camera would really benefit a new use-case. That’s where InDro Commander (and InDro Cortex, which we’ll touch on later) come in.

Pretty much every robot we ship comes with a version of InDro Commander or the more compact InDro Backpack. Both of these boxes instantly transform any platform into a smart robot.

How do they do that? The simple explanation is that Commander and InDro Backpack are a Plug & Play hub that combines powerful compute and remote teleoperation mission planning capabilities via InDro Controller with instant sensor integration. There are a couple of features here that are key.

InDro Controller is a secure, user-friendly remote mission planner. It allows you to plan repeatable tasks on complex missions with a few clicks. Want to zoom in on that gauge to ensure normal operating parameters aren’t being exceeded? No problem. Manually carry out that operation just once and InDro Controller will remember it for next time. (Oh, and InDro Controller is regularly updated.)

But the real key is Commander’s future-proofing. Clients can add sensors simply by plugging them into one of Commander’s multiple USB slots. The onboard Robot Operating System library will automatically detect the new sensor and add it to the InDro Controller dashboard. Clients don’t need to do any coding or soldering to ensure the perfect fit. (Plus, we have multiple autonomy software stacks, also continuously updated, that can be added.)

The ease of sensor integration with InDro Commander and InDro Controller means the client can modify or otherwise update their robot with additional or newer sensors without the hassle – and time and cost – normally associated with the task.

Below: InDro’s Commander – ready for anything you can throw at it

BUY QUALITY

You know that old maxim: You get what you pay for.

That’s usually true. But sometimes, as we’ve all experienced, you can overpay. And this is particularly true in the world of robotics. With some LiDAR units costing many tens of thousands of dollars, you’ll want to be absolutely certain you’re getting the best value (as with any other sensor). But with so many choices out there, how is a customer to discern what, truly, is ‘best’? It’s not like there’s a Rotten Tomatoes for LiDAR out there.

In the case of InDro, we’ve been building custom robots – for ourselves and our clients – for a long time. We’ve integrated and tested sensors from all the major manufacturers. And we’ve learned a lot during that process.

“We’ve built so many robots – and we’ve used pretty much every available LiDAR, depth camera, etc. on the market along the way,” says Corbeth. “As a result, what we offer in terms of components is the direct result of what we’ve seen have the best performance per dollar. We also look at: What’s easiest to work with from a development perspective? These criteria are all factored into our decisions on who we partner with.”

While Luke isn’t a trained engineer, he has a deep understanding of client needs, available solutions – and engineering in general. But when he’s got a question on behalf of a client, he also has our core R&D engineering team to tap on. We also have InDro Forge, our fabrication and manufacturing facility in Ottawa, for building custom enclosures and integration. This allows us full quality control, infinite customisation capabilities, and speeds the process of getting a finished robot out the door without having anything done by a third party.

AFTER THE SALE

InDro has built a reputation not only for its robots, but for its support of everything we sell. If clients ever have an issue, our engineering team gets to work. We have three well-established tiers of support, shown here in order of escalation:

• Remote troubleshooting
• Dispatching a Field Engineer to the client’s location
• Return to the robot to InDro for repair

“I would say in 90 per cent of cases we can we can remotely solve any issues – especially if it has InDro Controller. That makes it really easy to troubleshoot and update,” says Corbeth.

“But if it doesn’t, then sometimes it may make sense to go in-person and do those changes on behalf of the client on site.”

CORTEX

Because we’re a R&D company, we are constantly looking for ways to enhance our offerings. And while Commander is still both current and powerful, we’ve put considerable resources into making it even better – and much smaller.

InDro Cortex is the next generation of Commander, in a form factor so small it can easily be added to any platform: Wheeled, quadruped, humanoid – even a drone. With powerful compute, ROS1 and ROS2 libraries, InDro Controller and USB slots for sensors onboard, it’s ready for any mission. And, like Commander, InDro Cortex is a way of future-proofing your robot. If you’ve opted for InDro Controller or one of several autonomy stack options, you’ll always be working with the latest version of the software (in addition to being able to easily integrate new sensors).

“Cortex is the natural progression of Commander,” says Corbeth. “But because of the way it’s structured, it’s going to cost a lot less. So we’ll be able to outfit a robot with compute, sensors and connectivity at a fraction of the cost – which is really exciting.”

It is. And we look forward to a full commercial release of Cortex (which will also be available as a standalone product) by early summer. You can have a sneak peek, for free, below:

INDRO’S TAKE

If purchasing a robot is on your to-do list, we’d obviously like for you to consider InDro. But even if you don’t, we hope you’ll look for a supplier who offers everything we do, including:

• Track record with satisfied customers
• Components that have been field-tested for quality and value
• In-house manufacturing expertise
• Modular integration ease with after-sale support

The addition of InDro Forge to our capacity cannot be understated. We are now capable of all aspects of production, including IP-rated custom metal work for enclosures.

“Buying a robot is generally a significant capital expenditure,” says InDro Founder and CEO Philip Reece. “Our background as a research and development company, where we have invented and built so many robots from scratch, has taught us how to produce the absolute best products we can for the money. Those learnings – and there have been many – now benefit our clients.”

And, for those clients who know precisely what components or platforms they need, there’s also the new InDro Store. Happy shopping!