Drones these days get a bad rep. Guess that’s hardly a surprise when every headline is about another drone swarm taking an additional batch of human lives.
That’s not to say the disproportionate attention to the military use of drones is entirely fair. Because, within industrial settings, drones have propelled operational efficiency to a level none of us thought possible while sparing countless lives in the process.
On behalf of like-minded service providers who revel in the potential of drones to serve as an agent of the greater good, we hereby chart out all there’s to know about autonomous drones and how they’ve transformed our lives for the better.
What Makes a Drone Autonomous?
A drone is an umbrella term for any aircraft that can fly without a human pilot onboard. Hence the reason why a drone is often referred to as an unmanned aerial vehicle (UAV).
Now, there are two ways to go about flying an aircraft without giving direct commands. It can be controlled remotely or equipped with intelligent drone systems that allow the drone to self-pilot a premeditated flight path. The latter needs to be the case for a drone to be considered autonomous.
Are All Autonomous Drones Created Equal?
Here’s a million-dollar question. You see a drone zip through an obstacle course all by itself. Would that make the drone autonomous?
Well, not quite.
A considerable degree of autonomy is required to navigate a cluttered environment. Robust self-flying capacity, however, doesn’t make up for the whole story. For a flight to be truly autonomous, a drone must figure out when and where to fly, not just how.
Unfortunately, the level of technological maturity isn’t quite there yet. Whether it’s planning, dispatch, operational control, or postflight data analysis, a certain degree of human input is necessary to ensure the entire process remains airtight.
Levels of Autonomy
The chart above outlines different levels of drone autonomy. Let’s break it down a little more.
- Level 0. The pilot remains in full control throughout from takeoff to landing. Doomed to crash otherwise.
- Level 1. A specific function of the flight – generally position sensing or obstacle detection – is automated. Drones to remain in the pilot’s visual line of sight (VLOS).
- Level 2. Drones send out warning signs when an object moves in too close. The pilot is in charge of safe operation and keeping drones within VLOS.
- Level 3. The pilot is present purely as a backup. Payload comprised of devices and sensors allows drones to fly a pre-programmed path without running into obstacles.
- Level 4. On-site presence is no longer necessary. With clearance, operations can take place beyond visual line of sight (BLVOS) as drones are capable of flying through harsh conditions and have a failsafe to fall back on in cases of breakdown.
- Level 5. Drones handle all the legwork that goes into a flight. The technology and legal framework to accommodate full autonomy are still in the test phase.
Why Does Drone Autonomy Matter?
The degree of drone autonomy can make or break operational success.
Think about it. The whole point of using drones is to have them plow through and capture high-quality data from areas and assets that are either undesirable or unfit to be surveyed manually. With level 2 UAVs and below, the skill level of the operator will come to dictate how useful the collected data turns out to be.
Of course, this wouldn’t have been much of an issue had the talent pool of skilled pilots been deep. But tracking down an experienced and qualified operator who understands the structural complexity of the asset is like finding a needle in a haystack. And even when the right talent turns up, their services tend to cost all four limbs.
UAV Remote Sensing Applications
If you can’t already tell, the industrial use of autonomous drones is incredibly diverse–asset inspection, environmental monitoring, precision agriculture, site surveillance, and so forth so on.
One thing to note is that all these tasks fall under the roof of remote sensing, which, in essence, refers to contactless data collection.
So, how do autonomous drones go about remote sensing to collect aerial data? Three different ways.
Photogrammetry
Perhaps the most well-established aerial surveying technique, photogrammetry captures hundreds and thousands of snapshots over a specific surface area before stitching them together into a three-dimensional (3D) rendering.
The value of resulting 3D models is proving to be immense. Particularly within the O&M industry that has long been putting up with inefficient and hazardous manual inspection processes, drone photogrammetry is turning out to be a game-changer.
If anything, a video is worth a million words. Check out the clip below to get a close-up look into how drone photogrammetry takes place within the context of wind turbine inspections.
Light Detection and Ranging (LiDAR)
LiDAR won’t come to you as a brand-new concept should you be at all familiar with the workings of a radar.
As radar does with radio waves, LiDAR emits high-powered lasers at a target to measure the angle, intensity, and timing of the reflected pulses. The collected data points then go through mapping software to recreate a three-dimensional model, which helps site managers detect irregularities across a set area and gauge optimal uses of the land.
Keep in mind that the ability of LiDAR to provide a bird’s eye view of a mapped area doesn’t make up for the lack of photographic depth. To get the look and feel of a particular terrain supported by color and texture, you’ll need to opt for photogrammetry.