Unmanned Aerial Vehicles (UAVs) have come a long way. Early on, they were used for military purposes in World War II. In fact, some of the earliest models of drones were built here in Traverse City (home of Interactive Aerial) for a secret drone and pilot training program during the war. Today, drones are being sold in record numbers to both consumer and commercial buyers.The gateway to the skies opened in 2006, when the FAA put together rules that could make drones safe and practical for recreational and commercial use.1 That same year, the FAA issued its first commercial drone permit. Perhaps, around that same time period, the industrial sector started thinking about how to use robotics for visual inspections.
2005: "Is drone technology for inspections even possible?"
For years, businesses knocked around the general idea of using drones for inspections. It was certainly feasible that this technology could be used for inspecting hazardous and hard to reach areas. Presumably, an inspector would much rather fly a remote control drone than crawl around taking pictures in a boiler furnace. But there were many hurdles to jump through. Was the technology capable of navigating difficult areas and doing everything that a human inspector could do? Would it be difficult to identify and train drone pilots? Was the technology a sound investment? These were just a few of the questions companies were asking.
As popularity in drones grew, the ideas and plans started to coalesce and gather steam. As proof of the high level of interest, the FAA issued 1,000 drone permits in 2015.2 The permit rate increased further to 3,100 in the calendar year 2016.3 Once it was decided that drone inspections were a bonafide possibility, the industry transitioned to the next phase.
2015: "How can we prove that drone technology for inspections will work?"
Once the industry realized drone technology for inspections was possible, they turned to the next logical step—proving it. This couldn’t have come at a more critical time. In recent years, permit required confined spaces are still the leading cause of multiple fatalities in the workplace.4 In 2018, out of 100 confined entry deaths investigated by the Occupational Health and Safety Association (OSHA), 90 of them were due to either routine maintenance, repair, or inspection.5
For those reasons and more, many executives liked the idea of using drones for inspections, but they wanted proof that robotics could reduce risk, save time and costs, all while providing a meaningful and thorough inspection. This “prove it” stage has been happening for the past five years or so, and the entire industry has made amazing progress over this time period.
Today, there is significant proof that using robotics is a sustainable inspection alternative. The FAA estimates that there are nearly 1.6 million drones registered in the U.S. alone.6 Over 440,000 of those are commercial devices.7 For industrial inspections, the standards are being set. And now, almost a third of commercially registered drones are now being used for industrial inspections.8 And companies like Interactive Aerial are revolutionizing the way internal inspections are being performed. Now that we’ve made it through the “prove it” phase, what’s next? We believe the next big hurdle in the industry is going to be around repeatability.
The lead Boiler Inspector at a top U.S. power company recently told Interactive Aerial: “It took us 60-man hours to get the same pictures that your drone took in ten minutes.” While their team used a traditional personnel cart to inspect burners on one half of the boiler, our team inspected the other half with our Legacy One drone. Our drone photos were much more clear and consistent than the photos taken by inspectors in the cart. As a result, the company purchased multiple Legacy One drones and is now reducing its scaffolding requirements in lieu of newer robotic inspection methods.
Interactive Aerial Drone Outperforms Traditional Inspection Method
Current Day & Outlook
Repeatability is the Next Critical Step in Robotics Inspections
When we talk about repeatability, we’re primarily referring to the idea that inspection teams can quickly, easily, and safely compare defect areas on an ongoing basis. That means once baseline data is attained in a particular defect area, you can compare it through quantitative data, year after year. This instills confidence in knowing that inspections are comparing the exact defect areas examined in a previous inspection.
In the coming years, we think repeatability will transition through three levels—manual, automated, and autonomous. Each level builds upon the preceding one, making repeatability easier as it evolves. Today, manual inspections are happening with inspectors piloting drones by hand throughout a facility, which, looking at where we’ve come from, is a huge leap from traditional inspection methods, like building and crawling up scaffolding.
As we evolve to the next level, devices will be more automated and smarter and will literally be able to fly themselves.
Then, the ultimate level for industrial inspections are smart, AI-enabled devices that think completely on their own, performing autonomous inspections without the need for a drone pilot. The only human interaction will be setting the system up pre-inspection and then reviewing the vast amounts of data post-inspection.
The most important aspect of repeatability is the ability to go back and assess locations that have been identified in the past.
Whether we’re referring to manual, automated, or autonomous inspections, robotics no doubt makes repeatability easier. Human inspectors could feasibly follow repeatable guidelines, but it’s certainly not as easy or standardized, as using machines.
Here’s an example. Many large-scale assets might have thousands of identical components, or, areas to check. Look inside a boiler and it looks almost exactly the same everywhere you look, from the bottom hundred feet to the top hundred feet. When an inspector identifies a particular issue in the boiler, it’s critical they make note of the exact location of that defect.
If another inspector goes in a year later and doesn’t have that position data, that’s a critical gap. It’s this ability to go back, to inspect the same areas again to identify further deterioration, that makes repeatability so valuable. Robotics removes all these subjective decisions.
Organizations also need to be able to track the rate of growth of a defect from year to year. With things like photo measurement software, it’s now possible for devices to go back to the same spot and accurately calculate how much the defect has changed.
By making inspections repeatable and checking specific problem areas each and every year, you get a quantitative dataset with hard numbers rather than subjective opinions. You can identify the size of a defect at a given time Then, in later inspections, you can show how fast it’s growing and be able to make precise repair recommendations based on your company’s maintenance tolerance.
3 Levels of Repeatability
Let’s take a look at the three levels of repeatability that companies will encounter when it comes to the evolving nature of internal industrial inspections.
Manual inspections using robotics increases repeatability and reduces risk.
Inspectors today are fully capable of manually operating a drone or robot for inspections. This is a fairly new development, within the last decade. Manual inspections via drones helps companies in many ways, but first and foremost, it removes personnel from hazardous areas. This perhaps is the most important first step of this technology—keeping inspectors safe by enabling operation from outside dangerous assets.
A drone also provides information about the image it takes—such as the exact location of where it is inside the confined space area. Today, you can have the same drone, with the same camera, going to the same spots, year after year. This enables an organization to identify the rate of growth of a defect, and make budgetary plans for the next several years on how to repair it—which is not easy with different inspectors year after year.
The data is input to an advanced photo measurement software that provides quantitative data that can compare the defects and their rate of growth. An aspect of repeatability is the standardization of data being collected. With photo measurement software, the archiving of the data being collected is seamless, creating a digital footprint that can be quickly accessed at any time.
Meanwhile, the inspector doesn’t always have to go inside a confined space area. They operate the drone just outside of the area being inspected. The pilot has complete control of what to analyze. They are free to look at different areas multiple times if needed. Again, this is a valuable aspect of repeatability, the idea of checking specific areas multiple times to identify specific problem areas. However, manual inspections aren’t the “end all be all.” Sure, the human inspector is out of danger, but manual inspections are still labor intensive. And it’s still very difficult to manually fly a drone through a confined space environment.
Automated inspections using robotics make life easier for inspectors.
The next level of robotic inspections are automated capabilities. With an automated device, the inspector no longer needs to fully pilot the drone. It is automatic and performs its functions completely on its own. An automated drone is set up by the inspector, told where to go, but then proceeds to fly through the area taking pictures of everything it was told to do.
The inspector is no longer spending their time learning how to fly a drone or trying to maneuver a small device in hard to reach places. An automated drone is built for a specific area.
Think about a vending machine. It is an automated process. You want a candy bar, you see the candy bars there in the machine. You put money in, press a letter and number, and the candy bar automatically falls down into the tray. With automated inspections, the drone operator presses the right controls and, within the device, the function occurs through the right mechanisms.
In manual mode, the inspector has to manually fly to every spot, then snap an image with the right amount of overlap in each photo, ensuring that each picture is taken at the best possible angles, that the lighting is sufficient, and every inch of the site is photographed. With an automated device, all that is done for the inspector. Once the direction is given, it automatically performs all the functions it was told to perform. For obvious reasons, automated drones/robotics reduces operator stress, workload, and fatigue. After all, hundreds of hours of manually flying a device can be exhausting.
Automated inspections also provide great benefits when it comes to data collection. These devices take images of an entire asset automatically, providing a historical trend of information that makes data logging and archiving much more simple and thorough. The automatic process gives inspection teams fewer things to think about. They are ensured the actual inspection is complete, that every area of a site is photographed, while they can continue to focus on safety of the operation. Despite the improvement over manual inspections, automated inspections still involve an element of input from the inspection team. In other words, the inspector needs to tell the device where to go.
Autonomous inspections are the industry's Mount Everest.
We’re hearing a lot about driverless automobiles on the road, but autonomous robotics for inspections is gaining lots of momentum in the industrial industry. Autonomous devices are units that perform functions completely independently. They think on their own, and are self aware. Certainly, the inspections industry is not quite there yet, but there is no doubt that is where we are going.
Imagine an AI-enabled smart robotic device with its own sensors, navigation, geospatial awareness, multiple advanced cameras, lighting capabilities, memory, energy management, and ability to automatically analyze every inch of a site. Not only does it know where to go, but it learns along the way as well. Truly autonomous is all about self-awareness, recognizing areas it has been in before. It knows what specific areas to take pictures of, without any operator telling it what to do. In addition, and probably most important, it’s able to make decisions its own around safety and how to proceed when equipment failures occur.
Autonomous robotics beyond the capabilities of today's drone technology will transform the inspections industry.
Autonomous devices for inspections ensure the automatic repeatability of inspecting a particular site. It does this by removing the time, effort, and attention needed to ensure that repeatable inspections are being done. This outcome remains in the distance and is not entirely possible yet. But companies like ours are already working on autonomous technology for inspections every day. Just 15 years ago, drones were still a big question mark.
In the next 15 years, we will be working hard to evolve autonomous robotics into the next big thing.
Zenith & Repeatability
Zenith makes repeatability a reality for inspections.
At Interactive Aerial, we have already implemented the next level of inspection technology. When it comes to internal confined space inspections, our new Zenith Inspection System boasts many more advantages than today’s drones. Outdoor drones identify their height through barometric pressure and use compasses to also identify which way to fly. Inside a concrete structure, a compass doesn’t operate accurately.
Zenith technology is specifically built for internal inspections and is already performing repeatable and fully automated inspections. It uses a robust, military-grade inertial measurement unit (IMU) and highly accurate sensors to precisely measure the distance of the unit from ceilings, floors, and walls. This gives operators pinpoint accuracy of the depth and location of the device within the asset.
Additionally, with no complex in-person training needed, Zenith is operable by anybody, no certification required. It’s fully capable of being used as a one-man job for inspections all through online video training.
Zenith is built for today and for the future, with the ability to solve for every repeatability level along the way. As a manual & automated unit complete with photo measurement software, it currently supports an array of large companies in their annual inspection requirements. It will soon perform low-levels of autonomous inspections that will allow operators to focus more on the data instead of the inspection itself. While we solve inspection challenges of today, we’re always thinking about the future. This is why our Zenith technology is perfectly set up to evolve into a fully autonomous inspection unit in the years ahead. What can we say, we’re dreamers.
Learn more about Interactive Aerial’s Zenith robotics technology.
1"The History of Drones," Dronthusiast, 2018. https://www.dronethusiast.com/history-of-drones/
2"The Evolution of Commercial Drone Technology," Nevada Institute for Autonomous Systems, October 2018. https://nias-uas.com/evolution-commercial-drone-technology/
3"The Evolution of Commercial Drone Technology," Nevada Institute for Autonomous Systems, October 2018. https://nias-uas.com/evolution-commercial-drone-technology/
4"We Must Change the Statistics of Confined Space Injuries and Fatalities," Occupational Health and Safety Association, August 2018. https://ohsonline.com/articles/2018/08/01/we-must-change-the-statistics-of-confined-space-injuries-and-fatalities
5"We Must Change the Statistics of Confined Space Injuries and Fatalities," Occupational Health and Safety Association, August 2018. https://ohsonline.com/articles/2018/08/01/we-must-change-the-statistics-of-confined-space-injuries-and-fatalities
6Unmanned Aircraft Systems - Forecast," Federal Aviation Administration (FAA), 2019. https://www.faa.gov/data_research/aviation/aerospace_forecasts/media/Unmanned_Aircraft_Systems.pdf
7"UAS By the Numbers," Federal Aviation Administration (FAA), March 10, 2020. https://www.faa.gov/uas/resources/by_the_numbers/
8""State of Drones in the U.S.," Philly By Air, May 31, 2020. https://www.phillybyair.com/blog/drone-stats/