If you're interested in a quad loaded with open source software right out of the box, take a look at the Sky Viper v2450 GPS. It runs about $150.
It runs full ardupilot code, has a ublox gps chip, a taoglas gps antenna, and the video board and transmitter software is open source.
It runs a tiny webserver that allows you to configure any of the ardupilot variables, and is compatible with most of the ardupilot ground control software.
Also, just as an FYI, check out hovership.com - Steve doll runs that and he does some interesting things providing 3D printable components and kits, plus he’s just an awesome dude
Do you know how physically robust the Sky Viper is -- the article reports Bayangtoys X16 as being able to take a lot of beating; it would be great if Viper is robust as well.
Totally different classes, the bayangtoys is a brushless copter while the sky viper is a very lightweight brushed copter. Brushless will have an insane amount of power compared to brushed motors. Plus the ones on the sky viper are tiny.
The sky viper will hold well in crashes, you can pick up the non-gps version at Walmart. It's a kids toy so its made to be durable. That being said..if you're crashing your GPS assisted quadcopter you're already doing it wrong. I'm not really sure what applications a brushed GPS assist copter has though. It isn't powerful enough to carry any sort of payload and the camera quality is limited severely by that fact. Maybe useful for learning/tinkering with the code without the blender-ing capability that brushless motors have I guess.
Maybe I worded that incorrectly. What I mean is that you would be fine giving a kid a brushed quadcopter without worrying that they're going to filet themselves (for the most part. Yeah they could cut themselves but the damage would be order of magnitude less than with a brushless). Giving a brushless quadcopter to a kid without any instruction would be totally irresponsible. They're still toys yeah but can be pretty dangerous if you're careless.
I'm trying to understand the difference between the two types of motors and found this: https://lifehacker.com/are-brushless-cordless-tools-worth-th...
They talk about the more efficient battery use but not the added power or danger of brushless motors. Can you find a link that does explain it?
In short brushless motors are a 3 phase ac induction motor for the ones used in quads.
Brushed motors are the classic motors that you may have played with as a kid, put in a dc current and they spin.
Brushless motors are much much much more powerful and essentially only wear out due to heat, or bearings failing.
Brushed motors can be damaged by stalling, heat, etc and much more easily.
The important part that the above poster was trying to make is that a brushed motor, any size that would be used on a retail drone/quad/ etc might spin a prop with enough force to cut..... maybe even need a stitch or two but unlikely that bad.
Whereas the brushless motors used in even small quads would most certainly need stitches, and maybe even sever a finger.
Don't forget OpenTX, which is a transmitter firmware that overshadows any expensive as hell top-of-the-line brand RC transmitter by orders of magnitude in functionality, and runs on various mostly inexpensive hobby-grade hardware. Also Ardupilot (APM:Copter, APM:Plane, APM:Rover), PX4, and many many others.
I got a Taranis Q X7 that comes with OpenTX from the factory. I can't believe how high-quality both the firmware and the companion (cross-platform desktop app that simulates the entire transmitter) is.
After my last drone flew off never to be seen again I started looking into open source flight controllers. The open source platform has come a long way. I consider DJI to be the most advanced at FC firmware. But they intentionally limit you so much.
My next drones will most likely be a Spark for the interior and an easily upgradable OSS DIY rig for aerial photography. Running Betaflight or iNav.
I'd be interested to hear how the OSS community handle flyaway events. Every commercial drone has them, but they're tricky to code round being edge cases where the evidence has, well, flown away.
> Every commercial drone has them, but they're tricky to code round being edge cases where the evidence has, well, flown away.
I wonder, given how ultra small batteries and RF chipsets have become, if this would be a viable option: take a small battery, flash memory, a GPRS modem, a GPS receiver and something like the ESP32. Make a black box out of it by encasing it in epoxy to harden it against impact. Connect it via a data bus to the drone's main controller. When the power gets cut or the data stream from the main controller is interrupted or an accelerometer detects a hard landing, power up the GPS module and have it send the current location via SMS to the owner, and also use the ESP32's wifi capability to create an access point - which can then be triangulated using a simple mapper software on your cellphone that measures wifi strength.
On low price/toy grade quads, it is usually because they get out of range of the transmitter so you can't tell them to stop. If you check out some reviews on YouTube, you'll see the reviewer throttle up the quad and turn off the receiver to see how it deals with losing connection. I don't know enough about pricier drones to comment on why they would fly away - software glitch?
This is why I don't understand why companies lock down their hardware! You're getting free software enhancements and support; unlock that thing! Why waste people's time in reverse engineering your crap, when they could be spending that time building new functionality.
For example, Canon and the whole CHDK thing. Canon makes money from selling the hardware; so why lock down the software?
I work with hardware (not quadcopters) that is limited by software and the software is locked down, so I can at least tell you the reasons we do it:
1. Warranty. We know the limits of the hardware and our customers expect us to warranty that hardware for a reasonable amount of time, so we can't have consumers pushing the hardware beyond the duration/temperature/speed/etc that is going to significantly reduce its lifespan. We're not trying to be jerks, we're just saying here is our warranty and these are the parameters/limits within which we can offer that warranty.
2. Hackers. Hobby hackers are cool and most of us that work on these products are hobby hackers too, so we can appreciate the curiosity, fun and legitimate utility of modifying the software. But, there is a rising threat of malicious hackers and many of our customers (and non-customers) expect us to protect society as a whole from our products being taken over by malicious hackers for the purpose of harming someone. If nothing else, nobody who makes the products want to find out their product was used for that from a moral standpoint, not to mention a liability standpoint.
As for Canon, I'm sure people have tried to return camera bodies they've tried to hack and have bricked or otherwise damaged and that's not fair to Canon either.
The right solution would be for the manufacturers to make some of their software open source so hobbyists could add features and submit official pull requests that can be vetted by the engineers that are responsible for the reliability of the product and the safety of the people around it.
One reason, mentioned, is in order to reserve premium features.
Others are to maintain control over user experience; to simplify QA (making it less expensive, and leading to faster release cycles); reduce support costs; and increase engineering flexibility, by reducing the size of the API surface that requires a deprecation policy in order to avoid disrupting users.
So what exactly was done to the drone and how?
As far as i understood the article lists a few methods which could be used to mod the drone but not what he really did and how.
Has anyone managed to add automated/wireless charging to an inexpensive drone? It looks like there are a few commercial solutions out there, but nothing for less than $500.
Its sad that the FAA has rules that don't allow any commercial use of a drone in populated suburbs. With real time insurance starting to take off. These rules ought to be revisited, at least for light weight drones < 5 pounds.
Why are there no robotic airships available? They'd solve most payload and airtime issues that drons currently have. Also great for video since they're not as noisy as a quad and you could also turn off the engine.
In addition, size. While a drone can navigate flying in a busy street with all kinds of overhead wires, the huge size required for an airship would prevent many usage cases - the thing would simply get entangled in wires.
Airships are the best solution if you need something to stay in a specific area/position for longer times... something like an emergency cellphone/wifi BTS after major disaster strikes, but for moving around? Not really.
It runs full ardupilot code, has a ublox gps chip, a taoglas gps antenna, and the video board and transmitter software is open source.
It runs a tiny webserver that allows you to configure any of the ardupilot variables, and is compatible with most of the ardupilot ground control software.
http://sky-viper.com/product/info/V2450GPS-sd
http://ardupilot.org/
https://github.com/SkyRocketToys