Live data from Hacker News

Lily – Drone camera

news.ycombinator.com

411–420 of 496 posts

Re: Lily – Drone camera

#414

Earlier quoted context omitted.

As someone who has flown RC drones and heli for the better part of 5+ years now I generally find that there are 3 camps: 1) people who have never flown and assume the worst and often want unrealistic "guarantees" (see above), 2) those who fly, fly often and fly as safe as can be "guaranteed", 3) those who fly in an unsafe manner due to inexperience or just general carelessness. I've always tried to fly responsibly at…

But you do require a license to drive around a car, while this "hobby" does not require anything of that sort. Which is why it is questioned a lot. I belong to camp 2 as well, but i can understand why it might be a cause of concern for many.

I have a drivers license. I can drive a car. However, even I know I'm not the best driver in the world. In fact, I'm probably below average. I just drive so little.

A drivers license is a bad example.

Re: Lily – Drone camera

#415

Earlier quoted context omitted.

This is the big reason people don't buy these types of drones (this one certainly isn't the first of it's kind). It doesn't have obstacle avoidance or it will make a massive deal of it.

maybe someone can pull it off with ultra sound? Those can be pretty cheap.

Probably won't have much resolution for leaves. I've used ultrasonic transducers for a college class project almost 15 years ago. Ultrasound worked great for hard surfaces like walls, but softer materials like clothing reduced its range of detection. Snow covered trees, I think will have a similar signature that would be hard to detect except at close range. With a UAV, it could be done, but it'd have to be moving pretty slow I think. You'll probably have better luck with some sort of optical system.

Re: Lily – Drone camera

#416

I've been designing, building and flying model airplanes and later helicopters since I was ten years old. I've always flown in the context of a registered flying club and carry one million dollars insurance for my activity. I continue to be horrified by what is happening with these multicopters. People fly them outside of controlled environments and with a total lack of consideration towards others or their property.…

While acknowledging the fact that drones being operated by "morons" do not amount to safe. Could you fill me in on what safety features a 3k copter offers over one of these smaller and arguably less threatening drones. Googled a couple "high dollar" copters most seemed to be in range of ~4kg range with large fixed carbon fiber blades as opposed to (in this particular case) a 3 pound object with collapsible polycarbon…

I think you missed my point. Real model flight enthusiasts will only fly their birds at RC clubs. Clubs impose structure, safety rules, etc.

In other words, even if you are flying total crap the probability of hurting anyone or causing property damage is greately reduced.

To anser the other part of your question. A $3K heli has significantly better electronics and mechanics when compared to something like a DJI quad. And, of course, there's the fact that it has excellent autorotation with full control on the way down.

Yet, again, the point is we don't fly over people, roads or buildings and we have far more experience flying than most drone buyers.

Re: Lily – Drone camera

#417
post #46

Earlier quoted context omitted.

I can't speak for the Lily blades, but I have a couple of drones a little larger and the blades are harmless - they stop as soon as they detect obstruction and they are flexible/not sharp. It won't cut anyone.

I have NEVER seen a controller that would "stop as soon as they detect obstruction" like you say. In reality the flight controller is likely to do the opposite. Once it detects that the current speed of the blade is not enough to provide the desired lift, it will INCREASE the speed. Telling people the blades are harmless is very irresponsible.

The electronic speed controllers we use in RC have a loop time of a few milliseconds and are fully programmable. Some of them are capable of sensing the current going into the motor. You can also monitor the voltage instead of/in addition to the current. It's easy to cut the throttle when there is an unexpected power surge.

I personally haven't seen this technology in DIY multirotor parts, but I know that the AR.Drone has what they call "cut-out detection". They use it to prevent damage to the craft rather than humans, but the technology could be tuned for this as well.

Re: Lily – Drone camera

#418

I applied to YC with a speech interface for drones, one application of which would be hands-free operation of selfie drones like this. E.g., if you're mountain biking you probably don't want to take one hand off the handlebars and try to press the button on the remote that means "Switch to leading shot". Instead, you'd just talk into your earbud: "Switch to leading shot" or "Climb to 100 feet", or more complex cinema…

Interesting idea. I think the biggest challenge would be to filter out noise (i.e wind blowing on the microphone). But in a limited domain such as this you can do aggressive filtering and get fairly good results in very harsh environments.

Really cool idea. Now I have a strange urge to break out the ArduCopter.

Re: Lily – Drone camera

#419
post #121
post #37

Earlier quoted context omitted.

Somebody should do a spoof video of their marketing video that shows exactly that.

Somebody should make Office Space 2. http://www.imdb.com/title/tt0151804/

HBO's Silicon Valley series is probably the closest thing we're going to get to Office Space 2. Which is good, because the series is genius

Re: Lily – Drone camera

#420

Earlier quoted context omitted.

I was walking down Hollywood Blvd one day when they suddenly shut the street down for about 5 minutes to do a fly-by RC helicopter shot. This was a few years before lightweight drones and cameras were on the market and this helicopter reminded me of a flying lawnmower! As it whizzed by me and thousands of other people, I was pretty sure it could easily have taken off a limb. I have a fairly cheap drone with a built-i…

I believe NYT carried an article last year about an RC chopper flyer who got beheaded by his own chopper's main rotor. It happened in the Central Park. It was a large RC Chopper and something went wrong and he ended up get killed. He apparently liked to do some dangerous stunts apparently... However I would think these drones' propellers are shorter thus less dangerous? Any ideas?

He was flying a Trex 700-class stunt helicopter, with a large 1.6m diameter single rotor. He was using it for complex aerobatics - high accelerations, very high turn rates, very little respect for the gravity vector.

You can't do these sorts of things at all with human-sized helicopters - they're not maneuverable enough. In fact, the increase in maneuverability as high-powered single-rotor helicopters scale down is actually a problem for human reaction time. Model helicopters conventionally used a mechanical flybar in order to make the helicopter less responsive to inputs and more controllable. More recent innovations in flybarless electronics have enabled fly-by-wire control to substantially speed up responsiveness while still permitting stable flight when desired.

A single-rotor stunt helicopter is very different from a camera multirotor. A 1.6m carbon fiber blade vs a (usually plastic) 0.1-0.4m blade. A high-powered two-stroke engine vs a battery and motors.

And very importantly, the purpose of the two aircraft is entirely different. This type of helicopter uses collective and cyclic pitch oscillations, and a rudder. That's a completely different control scheme than a normal multirotor, which uses differential thrust. It lends itself to storing energy in the angular momentum of the blades at low collective pitches (running the rotor progressively faster while hovering), and then releasing it all in a maneuver by setting the pitch higher. That means the flat, strong carbon fiber rotor blades can be travelling at a tip speed of 400mph or higher. Even a very large camera-toting quadrotor is likely to have a (plastic, thin, bendy) rotor tip speed in the 100-150mph range - they're optimized for hovering battery life, and differential throttle has limited maneuverability, and the slower & larger the better. Being so much ligher per (disposable, easy to break) prop means that even at similar tip speeds the angular momentum is much diminished.

"Video footage has emerged of Mr Pirozek, who was a world-recognised aerobatic flyer, putting his Trex 700 helicopter through a series of remarkable tricks, including one that involves dropping the $1,500 model out of the sky by turning off the engines and restarting them just before the model chopper hits his head."

https://www.youtube.com/watch?v=J0wvXoZ9KYg

Lastly, "beheaded" seems to be a linkbait overstatement, and is considered dubiously by the medically inclined - even with all the power behind those blades, the skull is a very strong thing. The eyewitness comment was that there were wounds to his scalp, which tabloids turned into "cut off the top of his head" and then into "cut off his head". There were wounds to his scalp, but the ones to his throat are likely the ones that killed him. You don't need to sever the spinal column to slit a throat with a sharp blade.

Post reply on HN