Hacker News Reader

Comment by c_o_n_v_e_x | Hacker News Reader

I worked for a drone start-up that focused on a drone in a box / dedicated landing pad type solution. Seeing the big names and well funded competitors not succeed validated the technical and regulatory challenges for me.

The biggest technical challenge by far was getting the drone to land back in the box reliably over 10's,100's,1000's of attempts. Other annoying issues included buggy open source flight controller firmware, motor bearings only rated for 50 hours of flight (industry has since improved), and attempting to recharge batteries in hot ambient temperatures.

View on HN

I always thought the biggest impediment to drove dileveries would be a shotgun blast in certain parts of the country.

I guess multiple cams, and very quiet drones, would solve the problem.

I'm actually for drones. The sound of delivery vans delivering one package seems wasteful? Unless you are the driver?

Oh, there's no shortage of possible hazards for drones. Ever see a hawk take one down?

If the cameras are streaming back video that would be a very easy legal case against the person shooting. I'm guessing after it happens a few times and people see the repercussions they would stop.

Or a a legal case FOR the person shooting. Having a floating spy cam in my yard is shotgun worthy.

If you requested it and were informed of when and where from it was coming could wouldn't be able to make that argument.

This could be an interesting case - how high do property rights extend into the sky above your land? Obviously planes are allowed to fly over, even small private planes like Cessnas that fly pretty low. Is it actually the filming aspect that makes it "illegal"? What if the guy in the Cessna has a GoPro pointing out the window? What if it's not regular filming but say thermal imaging, do the same rules still apply?

In the US at least, I beleive the current legal definition is 500 ft in uncongested areas and 1000 ft elsewhere. Above that is considered "public highway" in federal law.

US v. Causby is the only ruling I could find (that was 83 ft).https://supreme.justia.com/cases/federal/us/328/256/

I think there may have been one or two other--possibly state level--cases. It's pretty ambiguous but loose consensus seems to be that you probably have property rights up to 400-500 feet. On rural land, a drone at 50 feet would seem to be fair game for skeet practice.

This actually happened, I believe the guy was in the wrong because the drone wasn’t on his property or something. The details are hazy.

Assuming there is a camera pointed in the right direction

Just like the all the successful legal cases against all those porch package thieves caught on amazons ring cameras.

I'm assuming that if you destroy amazons property they will be a lot more litigious than if you steal some guys headphones.

> people see the repercussions they would stop.

I doubt it. I think it's more likely that they'll do the shooting from cover so they can't be seen by the drone.

> I always thought the biggest impediment to drove dileveries would be a shotgun blast in certain parts of the country.

Bezos, in an all hands meeting was asked about that. "If we reach a point where people shooting down drones is our biggest barrier, I'll call the project a success".

What do you think of open source projects like iNav, Ardupilot, and PX4?

(I don't have enough open-space to test their navigation / RTH functionality, I use them only for poshold & althold in horizon mode)

For everyday users, I think they're OK. We were using Ardupilot and had very specific problems with the state estimation algos (kalman filters) for guidance, navigation, and control.

As part of our landing r&d, we tried RTK GNSS, which was generally OK in rural environments but a complete nightmare in built up urban environments. When the RTK GNSS receiver would lose fix, the system would fall back to standard GPS, thus causing a step change in position/state estimate of the drone, causing it to misbehave in flight. Fighting the "features" of the wrong tech is a slog, but I was not CTO... ce la vie.

Ultimately, we wound up licensing some tech for vision based landing.

That's fascinating, thanks! I knew navigation and control systems would be complex, but when I tried writing my own transfer functions, I got a taste for just how complex it really is. I admire what the USAF was able to do in the pre-integrated circuit era!

I guess I'm not surprised GNSS is so tricky for the final approach. Wide open spaces work great with averaging GPS, but as soon as some bogus data comes in shit goes sideways fast. I just assumed the experts figured that out, but it sounds like not the case if vision-based is the solution at that point.

Cool!

New L5 GPS should be able to help, at least in terms of not being thrown off by receiving reflections off of buildings (multipath). We'll see!

Reply on news.ycombinator.com