How drone autonomy ushers in a new era of AI opportunities

Join executives from July 26-28 for Transform’s AI & Edge Week. Hear top leaders discuss topics around AL/ML technology, conversational AI, IVA, NLP, Edge, etc. Book your free ticket now!

[Editor’s note: American Robotics is a commercial developer of automated drone systems.]

Drones have been talk about a lot for two decades now. In many respects, that attention was warranted. Military drones have changed the way we fight wars. Consumer drones have changed the way we film the world. For the commercial market, however, drones are largely a false start. In 2013, the Association of Unmanned Vehicle Systems International (AUVSI) predicted a 82 billion dollars market in 2025. In 2016, PwC predicts 127 billion dollars in the near future”. But we’re nowhere near those forecasts yet. Why so?

Let’s start with the primary purpose of drones in a commercial environment: data collection and analysis. The drone itself is a means to an end – a flying camera from which to get a unique aerial view of assets for inspection and analysis, be it pipelines, dumps gravel or vineyard. As a result, drones in this context fall under the umbrella of “remote sensing”.

In the world of remote sensing, drones are not the only players. There are high orbit satellites, low orbit satellites, airplanes, helicopters and hot air balloons. What do drones have that other remote sensing methods do not have? The first thing is: image resolution.

What does “high resolution” really mean?

The high resolution of one product is the low resolution of another.

Image resolution, or in other words Ground Sampling Distance (GSD) in this case, is the product of two key factors: (1) how powerful your image sensor is, and ( 2) how close you are to the subject you are photographing. Since drones typically fly very low to the ground (50-400 feet AGL), the chances of capturing a higher image resolution than with aircraft or satellites operating at higher altitudes are substantial. . In the end, you run into physical, optical, and economic problems, and the only way to get a better photo is to get closer to the subject. To quantify this:

  • “High resolution” for a drone works at 50ft AGL with 60MP camera about 1mm/pixel.
  • “High resolution” for a manned aircraft servicelike now no longer exist Terravionis 10 cm/pixel.
  • “High resolution” for a low-orbit satellite service, alike Planetary Laboratoryis 50 cm/pixel.

In other words, drones can provide 500 times more image resolution than the best satellite solutions.

The power of high resolution

Why is this problem? It turns out that there is a very direct and strong correlation between image resolution and potential value. As the computer phrase often says: “garbage in, garbage out”. The quality and breadth of machine vision-based analytics opportunities are exponentially higher at the resolution a drone can provide than other methods.

A satellite can tell you how many wells there are in Texas, but a drone can tell you exactly where and how the equipment on those pads is leaking. A manned drone can tell you which part of your cornfield is stressed, but a drone can tell you what pest is causing it. In other words, if you want handle a crack, bug, weed, leak or similar minor irregularity, you need the proper picture Resolution do like that.

Putting artificial intelligence into the equation

After obtaining the appropriate image resolution, now we can start training neural network (NN) and other machine learning (ML) algorithm to learn about these anomalies, detect them, warn them, and even potentially predict them.

Our software can now learn to distinguish between oil spills and shadows, accurately calculate the volume of a stockpile, or measure slight deviations in the track that could cause a derailment.

American Robotics estimates that over 10 million industrial property sites worldwide are already using automated drone (DIB) systems, which collect and analyze 20GB+ per day per drone. . inside United States aloneThere are more than 900,000 oil and gas well pads, 500,000 miles of pipelines, 60,000 electrical substations and 140,000 miles of track, all of which require constant monitoring for safety and productivity.

As a result, the size of this opportunity is really hard to quantify. What does it mean to digitize the entire physical assets of the world every day, across all major industries? What does it mean if we can start applying modern AI to petabytes of super high resolution data that has never existed before? What efficiencies are unleashed if you can detect every leak, crack and damaged area in near real time? Whatever the answer, I would bet the $82 billion and $127 billion numbers estimated by AUVSI and PwC are really low.

So: if the opportunity is so big and clear, why have these market predictions yet to come true? Enter the second important ability unlocked with autonomy: image frequency.

What does “high frequency” really mean?

The useful image frequency ratio is 10 times or more what people initially thought.

The biggest performance difference between automatic drone system and pilot is the frequency of data collection, processing and analysis. For 90% of commercial drone use cases, a drone has to fly over and over again and over the same piece of land, day after day, year after year, new value. This is the case for the fields of agriculture, oil pipelines, solar farms, nuclear power plants, perimeter security, mines, grazing and dumps. When examining the entire operational loop from setup to data being processed and analyzed, it became clear that operating a drone manually is not just a full-time job. And on average $150/hour per drone operator, it’s clear that the burden of full-time operations across all assets is simply not feasible for most customers, use cases, and markets.

This is the main reason why all predictions for the commercial drone industry have so far been delayed. Imaging assets by drone once or twice a year has little or no value in most use cases. For one reason or another, this frequency request was omitted, and until recently [subscription required]Automated operations that enable high-frequency drone testing have been banned by most federal governments around the world.

With a fully automated drone system, people on the ground (both pilots and observers) have been removed from the equation, and economics has completely changed as a result. DIB technology enables continuous operation, many times per day, at a tenth of the cost of a manually operated drone service.

With this increased frequency comes not only cost savings but, more importantly, the ability to track when and where problems occur and properly train AI models to do so autonomously. Since you don’t know when and where a methane leak or rail crack will occur, the only option is to scan every property as often as possible. And if you’re collecting that much data, you’d better build some software to help filter out important information for the end user.

Link this to real-world applications today

Automated drone technology demonstrates revolutionary capabilities in digitizing and analyzing the physical world, improving the efficiency and sustainability of our world’s critical infrastructure .

And thankfully, we have final move out of theory and into operation. After 20 long years of flying drones up and down the Hype Gartner Cycle, the “productivity plateau” is taking shape.

In January 2021, American Robotics becomes the first company approved by the FAA to operate an unmanned aerial vehicle system out of visual line of sight (BVLOS) there are no humans on the ground, important milestone Unlock the first truly autonomous operations. In May 2022, this approval was expanded to include Total 10 sites across eight US states, signaling a clear path for national scale.

More importantly, AI software now has a realistic mechanism to grow and evolve. Companies like Inventory report is using automated drone technology to measure daily stock and inventory volume monitoring. The Ardenna Railway inspector software is available path to scale across our nation’s entire rail infrastructure.

AI software companies like Dynam.AI only one new market for their technology and services. And customers like Chevron and ConocoPhillips are looking to a near future where methane emissions and oil leaks are dramatically limited by using Daily checks from automated drone systems.

My recommendation: Don’t look at smartphones, look at Oilfield, railway yard, assembly groundand farm for the next data and AI revolution. It may not have the same circumstances and status as the “metaverse”, but industry metaverse can have a stronger impact.

Reese Mozer is the co-founder and CEO of American Robotics.


Welcome to the VentureBeat community!

DataDecisionMakers is a place where professionals, including technical people who work with data, can share data-related insights and innovations.

If you want to read about cutting-edge ideas and updates, best practices, and the future of data and data technology, join us at DataDecisionMakers.

You can even consider contribute an article your own!

Read more from DataDecisionMakers

Source link


News5h: Update the world's latest breaking news online of the day, breaking news, politics, society today, international mainstream news .Updated news 24/7: Entertainment, the World everyday world. Hot news, images, video clips that are updated quickly and reliably

Related Articles

Back to top button