- Sounds like a neat space to be in. Wishing you calm skies next month. I guess you'll have all the goodies (RFI detectors, thermal imaging sensors, etc.) to collect data? I wonder to what extent detection from a distance (discovery?) and investigation of faults can be automated. Hopefully using drones for good will be sufficient for a viable business model. We've enough surveillance as it is.
- We're collecting 61MP RGB, LiDAR (DSM DTM) and radiometric thermal for sensors
We're thinking of flying at ~150' AGL or ~100' ATO at over 20 mph to collect data when automated. There's trade offs between effective speed, localized navigation, and mission planning. It's just challenging to build fully automated systems, but generally speaking flying higher and faster is more efficient and safer (also helps with command-and-control links)
- At my job [0] we work in this exact space, but on the data analytics side of things. We build computer vision pipelines for corrosion, components, defects, etc. We process imagery from drones, helis and satellite.
This is a tough problem to solve and for a lot of operators ultimately comes down to cost.
We're based in the UK but reach out if you wanna chat!
- Happy to chat! We’re 100% focused on data collection, so partnering with folks downstream is right up our alley.
And yes-- certainly a tough problem to solve! There’s pattern we keep seeing in the utility inspection space where teams start with a great defect detection or analytics product, then try to verticalize by adding data collection, basically spinning up a small aviation operation inside a SaaS company. I'm sure you've seen this plenty of times. Our hope here is to make collection simple enough so that doesn't need to keep happening :)
I'll reach out!
- How does this get around the problem you mentioned with FAA line of sight regulations?
- There’s no real way around BVLOS regs right now. The real problem is BVLOS + need to recharge. If you only solve one, you get limited value. Distributed charging without BVLOS is kind of useless. If a human still has to be there to maintain visual line of sight, it doesn't make much of a difference if the drone can recharge itself in the field. BVLOS without distributed charging also hits a value ceiling. You’re still constrained by battery swaps.
So the solution here is a deep understanding of FAA BVLOS waiver processes + a drone/pad network that is actually scalable.
That said, this is another benefit of working with utilities. When you’re inspecting power line corridors, you’re often operating in what the FAA calls “shielded area.” In practice, that just means you’re flying close to infrastructure that other aircraft tend to avoid anyway. That makes deconfliction much simpler, which in turn makes BVLOS waivers more attainable under Part 107.
The bigger shift is Part 108, which is expected to be finalized relatively soon, and should lower the barrier to BVLOS at scale.
- Follow up question, why aerial drones? Couldn't crawler drones work and avoid FAA regulations? Though obviously would be limited in speed and where lines have been damaged, though perhaps they could be managed to go around the damage via remote control.
- Crawler drones are not a bad idea! They're just a bit slow and limiting. Also doesn't drive our long term plan (can't leave power lines)
It's also mechanically complex and there's many line & pole types that make it challenging to build for.
When we let go of distribution line charging, we considered this idea. Off some rough calcs we thought there would be lines that we'd get 'stuck' on so that was hard to overcome...
- No feedback, this just sounds really cool.
- Is the inspection data you're accumulating becoming a moat in itself?
- We don't need to keep the data - we'd rather sell it to other (qualified) folks to build useful tools on top of our data platform.
Think verifying hail claim insurance for farmers (no need to send a guy out there), or helping NOAA assist truck drivers with weather conditions on interstates etc. etc. (the list is very long)
Maybe we'll cover all that surface area ourselves, but that sounds pretty challenging.
A good proxy is satellite companies. They're typically just a API call for data, not full-stack collection + analysis.
For utilities, we'll probably stay vertical, so that we have our first 'application' to bring our data to market. We'll see.
Also - at scale we're excited to help build physical world models with our data. To be seen if this is something we'd do in-house.
We're super excited to help new startups build data that were never possible before with the extreme high cost of data collection :D
- [dead]
26 points by wweissbluth 2 hours ago | 11 comments