Hard Light Productions Forums
Off-Topic Discussion => General Discussion => Topic started by: achtung on February 02, 2012, 10:55:06 pm
-
http://www.youtube.com/watch?feature=player_embedded&v=YQIMGV5vtd4
Get your crowbars ready.
-
First Gordon Freeman mysteriously appears at the LHC, and now this?
It's a sign I tell you. A sign!
-
(http://i.imgur.com/pf1wN.gif)
-
that is epic. just needs razor blade rotor wings and world dominion is within my grasp.
-
Saw this earlier today, totally awesome. I like the sound they make, like large, angry bees. And they move like strike craft from Homeworld, which is also good.
-
I've had a crowbar ready since I was born
-
I've occasionally been helping out a bit with getting this aeroquad ready for flight. It's about maybe 3 or 4 times as big as those nano quadrotors.
(http://i1210.photobucket.com/albums/cc418/est3b4n/AD.jpg)
In case you're wondering, no, I have no Combine-esque intentions to conquer the world or suppress humanity with this.
It's maybe not quite in the same category as those in that video, as it's not a drone. Still, I'll be standing well back when we test fly it proper.
-
I'd like to take some credit for calling them out as manhacks first. Kthx.
[22:46] <achtung> Actually, have you guys seen this? http://www.youtube.com/watch?feature=player_embedded&v=YQIMGV5vtd4
[22:46] <achtung> I'm amazed, and I want one.
[22:46] <achtung> No, I want twenty.
[22:47] <achtung> Nano Quadrotors
[22:47] <redsniper> it's crazy
[22:48] <@HerraTohtori> swarm
[22:48] <@HerraTohtori> that can not end well...
[22:48] <@HerraTohtori> those things in formation look freaky
[22:48] <redsniper> ffffuuuuuuuuuuu
[22:48] <achtung> Those will be weaponized.
[22:48] <achtung> I just know it.
[22:48] <@HerraTohtori> cyloooons
[22:48] <redsniper> technology is advancing TOO FAST
[22:48] <achtung> Those will fly right into your ****ing window, and boom!
[22:48] <redsniper> we are play gods!
[22:49] <@HerraTohtori> achtung: nevermind that, they will cut your face wide open
[22:49] <redsniper> manhacks!
[22:49] <dsockwell> manhacks?
[22:49] <redsniper> manhacks
[22:49] <dsockwell> hacks mentioned
[22:49] <redsniper> get your crowbars
[22:49] <redsniper> put in some batting practice
[22:50] <dsockwell> oh manhacks
[22:50] <dsockwell> got it
[22:50] <redsniper> manhacks
[22:50] <achtung> manhacks
[22:52] <achtung> You can hear the cameraman giggle when it flips.
[22:54] <SpardaSon21> ****, manhacks
[22:54] <SpardaSon21> don't bother hiding in the sewers
-
Better hope they don't teach theoretical physics at that unniversity...
-
Really. :blah:
A university is the last institution I would expect to abuse the "nano" prefix.
-
"It just means 'small' right?" "Yeah yeah, that's right. Like an ipod nano."
-
University != university physics department. :P
-
this manned octocopter (http://hackaday.com/2011/11/02/heres-your-flying-car/) was on hackaday awhile back, its the large version with more rotors, theres also the deathtrap chinese version (http://hackaday.com/2011/09/23/octocopter-will-someday-kill-someone/). if your intrested.
I've occasionally been helping out a bit with getting this aeroquad ready for flight. It's about maybe 3 or 4 times as big as those nano quadrotors.
-snip-
In case you're wondering, no, I have no Combine-esque intentions to conquer the world or suppress humanity with this.
It's maybe not quite in the same category as those in that video, as it's not a drone. Still, I'll be standing well back when we test fly it proper.
what kind of inertial reference you using on that.
-
I didn’t actually build it, and I’m not really technology-savvy. So far I’ve only helped out with calibrating the ESC (throttle range). The IMU (I take it that’s what you’re asking about?) is made up of the following:
- ITG-3200 triple axis gyro
- BMA180 triple axis accelerometer
- BMP085 barometric pressure sensor
- HMC5843 triple axis magnetometer.
There’s also an Arduino Mega 2560 and an AeroQuad Shield v2.0 in there. The clear plastic cover housing the electrical components is actually a food container (my suggestion).
So far it’s only made a short (about 1 second or less) hop. Something to do with the heading lock caused it to drop back down. I think it was enabled when it was facing a direction different to the heading it was facing after placing it on the ground, so once it was airborne it tried to abruptly swing back to that original heading.
I find it to be intimidating enough when it’s on the ground with all the rotors going. I especially cringe when I think of it hovering at head or eye-level. If one of those props flies off… :shaking: :shaking: :shaking: Perhaps a protective circular shroud could be added.
-
If you're thinking about protective shroud, you could try to also make it increase the lift of the quadcopter. That thing with ducted fans instead of propellers has a chance to look very futuristic, sort of like Dragon gunship from Avatar.
-
Agh! The Sound! I really don't like bees! :shaking:
-
I didn’t actually build it, and I’m not really technology-savvy. So far I’ve only helped out with calibrating the ESC (throttle range). The IMU (I take it that’s what you’re asking about?) is made up of the following:
- ITG-3200 triple axis gyro
- BMA180 triple axis accelerometer
- BMP085 barometric pressure sensor
- HMC5843 triple axis magnetometer.
There’s also an Arduino Mega 2560 and an AeroQuad Shield v2.0 in there. The clear plastic cover housing the electrical components is actually a food container (my suggestion).
So far it’s only made a short (about 1 second or less) hop. Something to do with the heading lock caused it to drop back down. I think it was enabled when it was facing a direction different to the heading it was facing after placing it on the ground, so once it was airborne it tried to abruptly swing back to that original heading.
I find it to be intimidating enough when it’s on the ground with all the rotors going. I especially cringe when I think of it hovering at head or eye-level. If one of those props flies off… :shaking: :shaking: :shaking: Perhaps a protective circular shroud could be added.
pretty cool, im thiking about getting an imu for my rc heli. the only stabilizer it has is a 1-axis gyro for the tail rotor. ive seen premade imu boards in the $60-$200, but if you buy the chips and make your own boards you can save a buttload of cash. does take doing a little smd soldering but im sure i can figure that out.
-
Awesome!
Like souped up wasps. Scary :nervous: :shaking:
-
If you're thinking about protective shroud, you could try to also make it increase the lift of the quadcopter. That thing with ducted fans instead of propellers has a chance to look very futuristic, sort of like Dragon gunship from Avatar.
Maybe that could work, I've passed the idea about having a duct-like shroud on. We would need to think about how to attach it to the frame. My idea is to use some stiff wire bracing to support a plastic shroud around each rotor. At present the main thing is to see if it can lift and hover in a stable manner. And I suppose with ducted fans it could resemble the Dragon gunship. Here's a scale RC model of one: http://www.youtube.com/watch?v=DlXlPVBQBlg (http://www.youtube.com/watch?v=DlXlPVBQBlg). I quite like the Scorpion and Samson designs (RC Scorpion here (http://www.youtube.com/watch?v=5gwlMIHIs4U)).
pretty cool, im thiking about getting an imu for my rc heli. the only stabilizer it has is a 1-axis gyro for the tail rotor. ive seen premade imu boards in the $60-$200, but if you buy the chips and make your own boards you can save a buttload of cash. does take doing a little smd soldering but im sure i can figure that out.
Not sure how much it's all come to money-wise, but from what I gather no pre-made boards were used. I think this (http://www.scribd.com/doc/46485031/AeroQuad-Manual) is the guide that was used to build it. Its modularity allows you to add other sensors such as a GPS receiver or an XBee shield (to provide wireless control via a computer, but I doubt we'll go that far). There's also a small HD camera that could be added. The AeroQuad Updater list has loads of modules to download and try out.
Links for anyone interested:
http://www.aeroquad.com/ (http://www.aeroquad.com/)
Instruction wiki: http://aeroquad.com/showwiki.php (http://aeroquad.com/showwiki.php)
There's also DIY Drones (http://www.diydrones.com) - these are amateur UAVs so they have GPS. AeroQuad is more modular with the sensors, whereas these have the sensors integrated into their shields.
The AeroQuad site has its own store, but there are others such as Flyduino (http://flyduino.net/) and Quadframe (http://quadframe.com/).
MultiCopter calculator: http://www.ecalc.ch/xcoptercalc_e.htm (http://www.ecalc.ch/xcoptercalc_e.htm)
Some very aerobatic quadcopters here: http://www.hovership.com/ (http://www.hovership.com/)
....You know, after looking at some of these 'copters I am reminded of the flying guillotine device I saw in that Mythbusters episode. Ugh... I'm imagining an RC version of Grant's "Decapitron".
-
im kinda past the point of stacking shileds. arduino is a good way to introduce newbs to mcu development or to provide dev hardware for people who just want to focus on the software side and dont mind spending money on shields. i myself find shields a little bit cumbersome for r/c projects. stack 3 or 4 of them and you got a massive pile of electronics and associated weight to deal with. the arduino board also kinda limits your mcu selection. sometimes you want to reduce the project size. maybe use a smaller mcu like a tiny 85 or 84, where you can do with fewer pins. for example if all your sensors and transceivers are i2c or spi, then you can really do without a lot of io, and a smaller mcu would be preferable. you may also want to do away with unused parts and connectors to reduce weight and cost (connectors are ****ing expensive).
the most expensive part of electronics manufacture is probably making pcbs (most parts cost pennies). using my zero budget mentality to hardware dev, i found a recipe for etching solution (being unable to acquire ferric chloride on the cheep), and another recipe for making your own transfer paper (yes you can use photo paper, but that costs actual money). then using a laser printer i got for free i can do the thermal transfer method. i will probably use eagle cad, it lets you print to transfer paper. the only thing i spent money on were the blank uncut pcbs. then its just a matter of print, polish, clean, iron on, fix errors with sharpie, soak, etch, drill, clean again, and solder **** to it. granted you can pay people to make boards for you, but whats the fun in that.
for me im using 4 servos and an esc, these all have their own driver circuitry so i only need to generate control signals. in software you feed the data from the imu into a pid algorithm to generate control signals. of course for a heli there is a lot more calibration to deal with, and channel mixing is more involved. i will probably deal with that (and generate control signals) on another mcu all together so that the pid algorithm just has to deal with roll, pitch, yaw, and collective (blade pitch and throttle). another mcu will run the show, host the bus, read the sensors, communicate over the transceiver, and do pid stuffs. i really dont want to make it autonomous, but i do want to have an autopilot to fall back on. it would try to level out and then reduce speed till its hovering. it would work in 3 modes, off, on, and fly by wire mode. fly by wire would just kinda prevent you from doing anything stupid. but its just a curiosity right now, not an actual project yet, il have to make sure my pcb manufacture process works.
-
In case you're wondering, no, I have no Combine-esque intentions to conquer the world or suppress humanity with this.
But, you do have intentions of taking over a less technologically savvy universe right?
-
I saw the vid and remembered the initial burst of fear when I first saw a manhack before remembering that I was Gordon Freeman and that **** don't mean jack to the guy who made Xen his personal *****/shooting gallery. Then I remembered that I'm not really a battle hardened physicist with a crowbar with a higher kill-count than that of the Grim Reaper's scythe, after which I promptly shat bricks.
-
I swear I broke into a panicked spaz attack of crowbar swipes/SMG rounds whenever I encountered one of those in the sewers. Thank the Maker they became child's play after you got the Gravity Gun.
-
In case you're wondering, no, I have no Combine-esque intentions to conquer the world or suppress humanity with this.
But, you do have intentions of taking over a less technologically savvy universe right?
Were that to be even possible, it wouldn't happen unless I'm allowed near the transmitter.
Or perhaps I am an unknowing pawn working for some... Benefactor... of sorts? D:
Over the weekend we tried to (crudely) get a feel for how much lift it produces at a given throttle setting by holding it down with one hand and raising the throttle (not a technique I'd encourage). Each motor is calibrated to draw about 20-22 amps at full throttle and with it set at about halfway (about 10-11 amps) it felt as though it had enough lift to raise your arm. Looking at the data in AeroQuad Configurator we found that for some reason the No.4 motor is drawing more current than the others.
This has in part led us to look at purchasing some XBee components to provide in-flight data monitoring, and the ability to fine-tune the flight parameters. There's certainly enough space for the required parts, although I don't know how much they'd add to the weight. Current all-up weight is about 1.5kg.
However, overcoming the weight doesn't seem like it would be an issue. Those 3-bladed rotors produce more lift than 2-bladed ones, but this results in increased torque and hence more power sappage (also the cold weather would reduce battery life, so testing is being done indoors for now).
This is just a project/hobby I'm helping out with, albeit it's one that scares me a bit. But I do think the idea of civilian UAVs is quite cool. I've seen these things on TV applied in livestock monitoring and I've heard about using R/C aircraft to explore mine shafts (probably for assisting in search and rescue operations).
-
my chinese made trex clone produces an insane amout of lift. after the lipo pack i was using decided to up and die, i didnt have money to replace it, so i gathered up every nicd, ni-mh, and even a few non-rechargable aa cells, taped and soldered them all together into an equivalent battery pack weighing about 2 pounds. and taped under the skids. it took off and flew for about 2 minutes before i crashed it. brushless motors pack a lot of power in a tiny and lightweight package. that first video i linked to for the manned electric octocopter you wouldnt think it capable of flight at first glance. companies have been trying to use the tech for all kinds of maned flight applications, namely planes for pilots with sport class licenses, but the faa really doesnt want to see cutting edge technology in sport class planes, and would rather stick to the tried and true lawnmower engines.
older gas helis had weight issues and the engines were not all that powerful for their weight. plus you needed to carry fuel in addition to batteries (usually heavy nicd cells) for the electronics. servos were bulky and no way around it you need at least 4 in a heli plus another to throttle the engine for gas helis. and you could have a gyro, which were heavy mechanical units. these days you have little 9 gram servos, escs are tiny and light weight, a single power system for everything, solid state gyros and a receiver smaller than a matchbook, leaves considerable margin of error as far as weight is concerned. quad copters likely have even more margin for error. 4 lifting motors, no servos (just escs), and you do all your stabilization and calibration with electronics (a regular heli needs to be mechanically calibrated as well as electronically) and in software.
-
Darn FAA... almost seems like they exist to keep flight out of the reach of the masses, for fear that they might hurt themselves. Can't we just have some sort of crash course in the basics of flight and then require that all pilots pack some sort of chute? heck, for a craft like the ones in the vid, your chute could almost save the craft too.
EDIT: OK, so you most likely could save the craft:
http://brsparachutes.com/
-
i think its more out of fear of hurting people on the ground / doing property damage. faa is a very conservative agency, they dont take risks at all. they like tried and true technology and they frown on anything new or untested. i cant really blame them, once you get off the ground you are pretty much in a do or die situation. its fully ok to put your own ass on the line, but when you consider anyone that might be below you, its not ok to put them in danger. best way to fly without the faa breathing down your neck is to get an ultralight, which as far as i know the faa doesnt really regulate, aside from restricting flights to rural areas and under a certain altitude.
-
Darn FAA... almost seems like they exist to keep flight out of the reach of the masses, for fear that they might hurt themselves. Can't we just have some sort of crash course in the basics of flight and then require that all pilots pack some sort of chute? heck, for a craft like the ones in the vid, your chute could almost save the craft too.
EDIT: OK, so you most likely could save the craft:
http://brsparachutes.com/
Build yourself some sort of ultralight and stay in Class G airspace nowhere near an airport or populated area.
Pick up a FAR/AIM if you're really interested, it's not that hard to decipher.
-
my chinese made trex clone produces an insane amout of lift. after the lipo pack i was using decided to up and die, i didnt have money to replace it, so i gathered up every nicd, ni-mh, and even a few non-rechargable aa cells, taped and soldered them all together into an equivalent battery pack weighing about 2 pounds. and taped under the skids. it took off and flew for about 2 minutes before i crashed it. brushless motors pack a lot of power in a tiny and lightweight package. that first video i linked to for the manned electric octocopter you wouldnt think it capable of flight at first glance. companies have been trying to use the tech for all kinds of maned flight applications, namely planes for pilots with sport class licenses, but the faa really doesnt want to see cutting edge technology in sport class planes, and would rather stick to the tried and true lawnmower engines.
older gas helis had weight issues and the engines were not all that powerful for their weight. plus you needed to carry fuel in addition to batteries (usually heavy nicd cells) for the electronics. servos were bulky and no way around it you need at least 4 in a heli plus another to throttle the engine for gas helis. and you could have a gyro, which were heavy mechanical units. these days you have little 9 gram servos, escs are tiny and light weight, a single power system for everything, solid state gyros and a receiver smaller than a matchbook, leaves considerable margin of error as far as weight is concerned. quad copters likely have even more margin for error. 4 lifting motors, no servos (just escs), and you do all your stabilization and calibration with electronics (a regular heli needs to be mechanically calibrated as well as electronically) and in software.
On the subject of power sources this quad uses LiPo batteries (lithium ion polymer batteries; they're kept in an ammo box for reasons you're probably aware of). Given the power demands it'll encounter we're looking at about 10 minutes of flight, but then, this quad wasn't really built for endurance; more for robustness.
Looking at the XBee radio modems it seems that there are basically two types - 2.4GHz and 900MHz, with maximum ranges of 1.6 to 24km (about 1 to 15 miles) respectively. Sounds impressive (to me, anyway).
Darn FAA... almost seems like they exist to keep flight out of the reach of the masses, for fear that they might hurt themselves. Can't we just have some sort of crash course in the basics of flight and then require that all pilots pack some sort of chute? heck, for a craft like the ones in the vid, your chute could almost save the craft too.
EDIT: OK, so you most likely could save the craft:
http://brsparachutes.com/
I'd suggest maybe taking gliding lessons if you're willing to fly unpowered. I've read that pilots that have taken gliding tuition prior to flying engine-driven aircraft say they got more of an understanding and appreciation for the forces keeping them aloft than they would have had if they went straight into learning to fly with an engine. It helps add to the pilot's confidence, especially when faced with a dead-stick landing situation.
-
my chinese made trex clone produces an insane amout of lift. after the lipo pack i was using decided to up and die, i didnt have money to replace it, so i gathered up every nicd, ni-mh, and even a few non-rechargable aa cells, taped and soldered them all together into an equivalent battery pack weighing about 2 pounds. and taped under the skids. it took off and flew for about 2 minutes before i crashed it. brushless motors pack a lot of power in a tiny and lightweight package. that first video i linked to for the manned electric octocopter you wouldnt think it capable of flight at first glance. companies have been trying to use the tech for all kinds of maned flight applications, namely planes for pilots with sport class licenses, but the faa really doesnt want to see cutting edge technology in sport class planes, and would rather stick to the tried and true lawnmower engines.
older gas helis had weight issues and the engines were not all that powerful for their weight. plus you needed to carry fuel in addition to batteries (usually heavy nicd cells) for the electronics. servos were bulky and no way around it you need at least 4 in a heli plus another to throttle the engine for gas helis. and you could have a gyro, which were heavy mechanical units. these days you have little 9 gram servos, escs are tiny and light weight, a single power system for everything, solid state gyros and a receiver smaller than a matchbook, leaves considerable margin of error as far as weight is concerned. quad copters likely have even more margin for error. 4 lifting motors, no servos (just escs), and you do all your stabilization and calibration with electronics (a regular heli needs to be mechanically calibrated as well as electronically) and in software.
On the subject of power sources this quad uses LiPo batteries (lithium ion polymer batteries; they're kept in an ammo box for reasons you're probably aware of). Given the power demands it'll encounter we're looking at about 10 minutes of flight, but then, this quad wasn't really built for endurance; more for robustness.
Looking at the XBee radio modems it seems that there are basically two types - 2.4GHz and 900MHz, with maximum ranges of 1.6 to 24km (about 1 to 15 miles) respectively. Sounds impressive (to me, anyway).
lipos are nice, but we have only a few months worth of good flying weather every year, and ive yet to come up with a way to make the batteries survive the off season, so i end up buying a new pack every year. high power batteries like that have no control circuitry and arent as storage happy.
xbee is out of my price range (its not so much the price but that you need to buy 2 of them to do anything useful). they are nice units though, there is one with a 40 mile range if you have the cash. i had good luck with some 434 and 315 mhz ask modules, the cheap ($2-$3) ism ones that hardly ever work right. they have a really low baud rate (4800bps, sometimes), and i very much doubt the fcc approves me using them. ive been able to make them work well enough to build an rc car from junk/legos. i recently bought me a pair of 2.4ghz nordic (nRF24L01+) modules on the cheap ($7 on ebay) to play with, but i havent tried using them yet. they are rated for like 2mbit, and im not sure what kinda range they are. supposedly range ultimately depends on transmit power and antennae configuration. 2.4ghz has problems getting through objects, and is better for line of site applications (my heli radio is 2.4 as well). i dont expect this module to perform well at all. but im sure its better than what ive been using.
-
http://hackaday.com/2012/02/24/variable-pitch-quadrocopter-flies-upside-down/
now were in a world of ****
-
It certainly seems to change attitude and height in a more refined way. On that subject, the quadcopter I'm helping out with now has an ultrasonic sensor for more precise altitude monitoring. In the end we didn't bother with getting an XBee board but it might be something to add later on.
We could do with having some suitable indoor testing space, preferably with a nice level surface so the quadcopter lifts off along its normal axis. And some netting like in those videos would be a good idea.
Actually, I might get around to helping with the flight testing (hovering) tomorrow.
-
i was looking at my heli's aluminum tail rotor assembly, and it looks like i could easily re-configure them to spin in either direction while still producing positive thrust. so i bet if i had 4 of them i could mod them to provide prop pitch for each motor in a quad copter, or possibly run them all off of a single motor. of course each would require a servo (my chopper uses a 9 gram servo to actuate the tail rotor) to operate. im not sure what the yaw performance would be running off of a single motor. as i understand it a rotor's torque goes up with blade pitch, so you should be able to create the torque differential with blade pitch alone.
-
manhacks seen developing culture http://www.youtube.com/watch?feature=player_embedded&v=_sUeGC-8dyk
request hlp threat down update
-
manhacks seen developing culture http://www.youtube.com/watch?feature=player_embedded&v=_sUeGC-8dyk
request hlp threat down update
Cartographer UAVs and Anarchist Quadrotors?
On a semi related note I've noticed that when you turn off my mom's CRV it makes a dead on impression of the Combine Overwatch data squelch noise when troopers are terminated.
-
request hlp threat down update
Remind me tomorrow.
-
Done. (http://www.hard-light.net/forums/index.php?topic=69862.msg1588702#msg1588702) I did it for you, battuta. :p
-
more serious vid on the subject
http://www.youtube.com/watch?v=4ErEBkj_3PY&list=UUAuUUnT6oDeKwE6v1NGQxug&feature=plcp
-
more serious vid on the subject
http://www.youtube.com/watch?v=4ErEBkj_3PY&list=UUAuUUnT6oDeKwE6v1NGQxug&feature=plcp
The end is absolutely wonderful.
-
red sniper best sniper, also best color
-
It certainly seems to change attitude and height in a more refined way. On that subject, the quadcopter I'm helping out with now has an ultrasonic sensor for more precise altitude monitoring. In the end we didn't bother with getting an XBee board but it might be something to add later on.
We could do with having some suitable indoor testing space, preferably with a nice level surface so the quadcopter lifts off along its normal axis. And some netting like in those videos would be a good idea.
Actually, I might get around to helping with the flight testing (hovering) tomorrow.
Make sure you turn off the AC and any fans inside. Indoor flight gets hairy, because the recirculating air has a tendancy to significantly reduce lift at seemingly unpredictable moments. The nanoquads seem to be pretty sturdy in this sense, due to a couple of things.
Oh, btw. To my knowledge, bot formations and stuff like that has been around awhile. Just look at video games and simulations. Not sure how the swarm is controlled though, if it's controlled by a central computer or if they're doing it individually. (probably a combination)
-
I believe there's no central intelligence, just the flock. It'd be less impressive otherwise.
-
provided every quad copter had a set flight plan to be at a certain place at a certain time, and assuming the time between all quad rotors was in sync, you wouldnt need any external computer to control them. not to say that you cant do that arrangement, but it is certainly within the capabilities of a typical microcontroller to execute a well choreographed flight plan. if you want quad rotors that can analyze unfamiliar surroundings and navigate through them safely, youre gonna need more cpu power. of course with arm processors getting tiny as ****, its likely you could carry that cpu on-board and have some degree of autonomous operation, in say, an urban environment. having a quadrotor as a client attached to a sentient supercomputer would be a very bad situation indeed.
-
provided every quad copter had a set flight plan to be at a certain place at a certain time, and assuming the time between all quad rotors was in sync, you wouldnt need any external computer to control them. not to say that you cant do that arrangement, but it is certainly within the capabilities of a typical microcontroller to execute a well choreographed flight plan. if you want quad rotors that can analyze unfamiliar surroundings and navigate through them safely, youre gonna need more cpu power. of course with arm processors getting tiny as ****, its likely you could carry that cpu on-board and have some degree of autonomous operation, in say, an urban environment. having a quadrotor as a client attached to a sentient supercomputer would be a very bad situation indeed.
I don't believe this is simply preset choreography; I believe the rotors have individual navigational ability.
-
Absolute coordinates and the real world don't work very well together, so I highly doubt they're preprogrammed flight paths.
-
I keep on imagining a flock of them flying in formation and dragging a monomolecular net behind them, chasing after my enemies.
-
Absolute coordinates and the real world don't work very well together, so I highly doubt they're preprogrammed flight paths.
provided an adequate inertial and positional references. the cumulative data from a gyro, accelerometer, input from gps, magnetometer, barometric sensor, optical rangefinders, cameras, etc, are sufficient parameters for a close approximation of actual position with regards to the earth. the more data the better. this data is of course subject to some error, thats why you have other sources of the same data to compare against. you can loosely follow an established flight plan, and any error between the real flight path and the virtual flight plan can be compensated for with clever pid algorithms. and you would want to have some proximity sensors to feed collision avoidance routines.
so there is some intelegence for following the flight plan. of course having it talk to a remote computer adds some latency to data logging and response commands. and the sensory data would still be the same (though you may add external motion capture systems as an additional data source, which is likely something done in the videos).
provided every quad copter had a set flight plan to be at a certain place at a certain time, and assuming the time between all quad rotors was in sync, you wouldnt need any external computer to control them. not to say that you cant do that arrangement, but it is certainly within the capabilities of a typical microcontroller to execute a well choreographed flight plan. if you want quad rotors that can analyze unfamiliar surroundings and navigate through them safely, youre gonna need more cpu power. of course with arm processors getting tiny as ****, its likely you could carry that cpu on-board and have some degree of autonomous operation, in say, an urban environment. having a quadrotor as a client attached to a sentient supercomputer would be a very bad situation indeed.
I don't believe this is simply preset choreography; I believe the rotors have individual navigational ability.
its likely a little of both. its impossible to follow the flight plan accurately, especialy if the path is physically impossible to follow with the uav. i was really just trying to compare the computational facilities of various control systems in that post and what they can do in terms of navigational capabilities. mcu being the least powerful, lightweight arm computer being a bit better, exteranl computers, supercomputers etc, being able to do some really fancy stuff.
regardless of the processing power of the computer that is flying it, the sensor data input would be the same (and as i said earlier you might use some off vehicle sensors as well), and the meathods for output would be the same, the only variables are how much processing power do you have, and how far away is that processor or external sensor data (manifested as latency, which in this case is bad).
-
regardless of the processing power of the computer that is flying it, the sensor data input would be the same (and as i said earlier you might use some off vehicle sensors as well), and the meathods for output would be the same, the only variables are how much processing power do you have, and how far away is that processor or external sensor data (manifested as latency, which in this case is bad).
I think the flight controls itself are fairly simple, just a couple of multiplies and additions for a PID or FLC type system. Most of the computation time would then be processing and filtering the data from the sensors, with the CCD and image processors the most viscous (I assume).
-
yea, most sensors are low throughput devices. they dont need to move a whole lot of data, and so they can be sampled frequently and fed into the control algorithm with a minimum of latency, whereas it might take a microcontroller or a low end arm processor several hundreds or even thousands of milliseconds to process an image into a useful format for analysis. you can also throw hardware at the problem, one quad coptor had a ms kinect on top of it for realtime mapping of the environment. figure you could also use an fpga to accelerate the image processing to abstract the data coming out of the ccd (i think this is what trackir uses, the high pin count on the main chip seems to indicate its an fpga, though wiimotes seem to do the same job with a microcontroller just fine). really cpu power isnt all that much of a problem, via has a 6cm^2 x86 form factor and im sure you can get smaller with a high end arm processor. it wouldnt supprise me if you could get a full 1ghz without too much additional weight over a simple 8 bit mcu.
-
An fpga would be real handy for the sensor filtering and fusion, since you can literally allot a portion of the thing to different tasks, and might even be able to handle the 6 or so dimensional controllers on top of that.
Watched a video of a quadrotor presentation at one of the TED conference today, something that struck me as odd is that the speaker mentioned that the core did control calculations 600 times per second. I find this odd, since most commercial ESC's out there operate at 50Hz, and even if you turbo the PWM you top out at about 300Hz. Those rare I2C ESC's are wonderful from what I've read, but even they top out at around 300Hz (unless I missed something).
So, even if you did perform 600, or even 1000 or 2000 control algo passes, your not going to get a change in performance if your ESC can't keep up.
-
most of the hobby escs use hobby servo signaling for input, a 0.5 to 2.5ms pulse every 20 ms, where the length of the pulse length indicates the position of the servo (or in this case the motor speed). so roughly they can be updated 45 times a second. its ok, but not a lot. you have other delays as well. motors use coils, and coils are inductors, and inductors resist changes in current. so it takes a while for them to power up or down when the current changes (not long but log enough to cause some error). i figure its electronically impossible to do frequent updates to high power motors like these. thats why the quad coptor with the variable pitch rotors are more maneuverable than the ones those controlled with escs alone. even if your not regulating the motor speed that rapidly you can still sample the sensors and run your control algorithms so that when it comes time to update the motor's output you have the best solution possible to get the desired output.
the tricky part is that all your sensors may update at different rates, and all your control systems have electrical and mechanical delays, you probably cant sample your fps as fast as you can sample your gyro and accelerometer. any sensors that need to process a large amount of data will not be able to update very quickly. essentially all your ins and outs operate at different sample rates. you could run it all in one main loop, but you would be limited by your slowest sample rate. you kinda have to employ some degree of multi-threading so that all your inputs and outputs can be updated at their optimal rate. i dont want to even think about the kind of mayhem that causes on your control system.
-
The smaller motors may have an advantage on that note, since they not only have a lower moment of a inertia than the larger motors, but also their inductance per coil is reduced (since they usually have fewer turns per coil, and I think the smaller radius plays a role, too).
One interesting thing that I did notice from the TED video was that the guys related several key components, such as lift and rotational rate, to the radius of the propeller. I should point out though that this is not exactly the case, as they've forgotten about the pitch of the propeller as well.
For those who don't know, the fastest a propeller can ideally churn through any media is the same speed as its pitch. Realistically, I think that the propeller itself can achieve about 90% of its theoretical maximum, and less when the drag of the craft it's pulling (or pushing) comes into play.
I know that for a fact, in an [outrageously] ideal fluid, the calculations for the craft's maximal rotational speed is (w*P) / L, where w is the rotational speed of the motor ( unit revolutions per second), P is the pitch (unit distance per revolution) and L is the distance from the craft's center of gravity (same unit distance).
even if your not regulating the motor speed that rapidly you can still sample the sensors and run your control algorithms so that when it comes time to update the motor's output you have the best solution possible to get the desired output.
That's more or less oversampling, which should have the same effect as anti-aliasing. I think, however, that the ESC's play a role here too, as the one's I've used have a Nyquist filter on them (essentially it halves your bandwidth right then and there). Computations that have a relative frequency that is higher than the ESC's bandwitdth will be tossed out... meaning you just wasted a bunch of power on calculations that you don't use.
the tricky part is that all your sensors may update at different rates, and all your control systems have electrical and mechanical delays, you probably cant sample your fps as fast as you can sample your gyro and accelerometer. any sensors that need to process a large amount of data will not be able to update very quickly. essentially all your ins and outs operate at different sample rates. you could run it all in one main loop, but you would be limited by your slowest sample rate.
This is where frequency analysis and design comes in handy. You can apply frequency filter on the algorithms per sensor. A really basic example of this doubles as a simple way of fusing the IMU data of the accelerometer and the gyrometer: the complimentary filter. What this does is filter out the frequencies considered "noise" by both sensors and fuses them into one reading, say, your pitch angle. If you did your discrete math right, you'll get a clean reading from across both sensors, with either a small dip or a small bump in the signal gains at the points the sensor's bandwidths overlap.
you kinda have to employ some degree of multi-threading so that all your inputs and outputs can be updated at their optimal rate. i dont want to even think about the kind of mayhem that causes on your control system.
Multi-threading is definitely a must in this type of system, even though the control algo's themselves are pretty much pipelined, the sensor data will be used in multiple control algo's... Top that with sensors (EDIT: and/or algo's) operating at different frequencies and voila, a need for an OS.
Oh, and it shouldn't cause much mayhem with the control system, at least with a PID or FLC type. Outputs of some controller blocks will be an input to others (like linear motion down into rotatory motion), and they all finally get summed up to the final output.
-
ive really only done basic control systems, i tried some pid stuff (cnc oriented) with limited success.
in case anyoen hasnt seen the ted video:
http://www.ted.com/talks/lang/en/vijay_kumar_robots_that_fly_and_cooperate.html
thers some stuff in it that wasnt posted here.
-
it doesnt fly but its another badass weapon in the ai arsenol
http://hackaday.com/2012/04/02/sphere-morphing-hexabot-now-rolls-around/
-
Morph ball!
Fffff****! We are a civilization that builds robots! Not just lame automated welder bots, but actual crazy sci-fi **** robots!
-
Good lord, it even SOUNDS like a manhack. Run!
(Of course, what's easier to shoot out of the sky than something barreling straight for you at zero declination? So no, Gordon, put down the crowbar, and get your shotgun!)
-
it doesnt fly but its another badass weapon in the ai arsenol
http://hackaday.com/2012/04/02/sphere-morphing-hexabot-now-rolls-around/
Transform and roll out!
-
Wow, scale that up, color it MARPAT, add some weaponry and blamo instant stalker tank.
-
I am reminded of those rolling bots from Star Wars Ep1.. what were they called? Anyways, you know what I'm talking about.
-
I am reminded of those rolling bots from Star Wars Ep1.. what were they called? Anyways, you know what I'm talking about.
Droidekas.
-
I believe this is what you all have been waiting for: http://www.youtube.com/watch?feature=player_embedded&v=SNPJMk2fgJU#!
Enjoy :) LOL.
-
<3
looks fake though, that gun doesnt look like it would do any more than punch a few holes. they are using some pyrotechnics there to make it look more dramatic. on the other hand ii dont see any reason why this wouldnt work though. quad copters seem to have a really high lift capacity, and so long as the recoil from the gun is vectored into the cg of the copter, i dont see any reason why you cant arm them. people already put guns on r/c airplanes and have dogfight competitions.
-
vid is totally fake, concept is totally plausible.
-
Even though the vid's a fake, the casual drone hobbyists are having heart attacks from it, due to the very fear of losing their current freedoms of actually playing with them in non-threatening ways.
That, and it makes world domination oh so much harder to do. :P
-
:wtf:
If they outlawed drones for that, well, what if someone put a semi-auto set of rifles in a car? 4x front turrets, 2x each side, 4x rear. Attach controls to auto-fire the weapon (repeated trigger pull or modify the firing mechanism to do full auto) and you have a nice terror weapon. You can stuff a lot of rounds into a car if you create a custom clip design. How many casualties on the freeway before you got caught? Or in a parking lot or worse, one of those strip malls?
Therefore, let us ban automobiles!! :lol:
-
interstate 76 was awesome. it was like freespace with cars, and disco music. :D
disco sucks
-
:wtf:
If they outlawed drones for that, well, what if someone put a semi-auto set of rifles in a car? 4x front turrets, 2x each side, 4x rear. Attach controls to auto-fire the weapon (repeated trigger pull or modify the firing mechanism to do full auto) and you have a nice terror weapon. You can stuff a lot of rounds into a car if you create a custom clip design. How many casualties on the freeway before you got caught? Or in a parking lot or worse, one of those strip malls?
Therefore, let us ban automobiles!! :lol:
Believe it or not, but that idea's been around since the first James Bond movies, if not earlier. Therefore, it's not fashionable or very realistic to the paranoid, because it's widely known that you'll only ever see that happening in the movies or on TV.
Drones, on the other hand, are not at all common to everybody. Do you see a drone in everybody's driveway? No. Do you see a drone at every airport? No, (or not yet at least). Therefore, drone's currently boast the unknown factor that is very famous for instilling paranoia and fear.
Another thing against drones? They're machines. Even though people know that total machine takeover only happens in the movies, it's still a serious threat and is increasingly gnawing at the minds of everybody who fathoms their potential.
-
Kitty! (http://www.dailymail.co.uk/news/article-2154283/Cats-away-Artist-turns-dead-pet-flying-helicopter-killed-car.html)
-
making stuff animals fly is more disturbing to me than just hanging them on a wall somewhere. :eek:
-
speak for yourself, a deer head makes an excellent coat rack!
-
speak for yourself, a deer head makes an excellent coat rack!
Yes, but a flying deer head is just creepy.
-
speak for yourself, a deer head makes an excellent coat rack!
Yes, but a flying deer head is just creepy.
You sir, have obviously never seen 'Return To Oz'. Okay so it's a moose head, but they're practically the same animal.
-
That movie was ****ing weird.
-
That movie was ****ing weird.
And yet, if it were on TV right now we'd all still probably put it on.
-
So now they're conducting lightshows (http://www.youtube.com/watch?v=cseTX_rW3uM&feature=youtu.be).
-
yep... still gotta get my OS finished.
-
oops, I'm dumb.
-
thread read fail is fail
-
Did any of you, by chance, flew an RC octocopter in Geneva, Switzerland about a month ago? :)
-
Did any of you, by chance, flew an RC octocopter in Geneva, Switzerland about a month ago? :)
No, saw a guy flying a hexacopter for the TXWS a few weeks ago. Very impressed how smooth it flew.
-
Not my thread, I know, but my justification for necroing is that I've posted in it before and would like to continue with some more relevent content, as one or two people here may find it of interest.
My elder sibling managed to give the AeroQuad (AKA the flying food container, in this case) its first proper flight about a month ago. I went along and took some photos and made some footage.
It now sports a wider, more sturdy set of landing skids, and a small 720p camera, which is the small black object located just beneath the blue clasp of the container facing you in the second pic.
(http://i1210.photobucket.com/albums/cc418/est3b4n/5fd55600-2e27-4e4d-b89b-a3612e94fb00_zps14ca0b1d.jpg) (http://s1210.photobucket.com/user/est3b4n/media/5fd55600-2e27-4e4d-b89b-a3612e94fb00_zps14ca0b1d.jpg.html)
(http://i1210.photobucket.com/albums/cc418/est3b4n/6637c7a2-8de2-40b1-81ec-6f36495c34bd_zpsa8f416f0.jpg) (http://s1210.photobucket.com/user/est3b4n/media/6637c7a2-8de2-40b1-81ec-6f36495c34bd_zpsa8f416f0.jpg.html)
Lift-off.
(http://i1210.photobucket.com/albums/cc418/est3b4n/AQ_lift-off_zps542a6fd6.jpg) (http://s1210.photobucket.com/user/est3b4n/media/AQ_lift-off_zps542a6fd6.jpg.html)
(http://i1210.photobucket.com/albums/cc418/est3b4n/AQ_ascent_zps40396350.jpg) (http://s1210.photobucket.com/user/est3b4n/media/AQ_ascent_zps40396350.jpg.html)
On the first flight, it was taken up to about maybe 50 or 60 feet at its highest altitude, but without a suitable reference I'm not absolutely certain. Here I think it was at around 20-30 feet.
(http://i1210.photobucket.com/albums/cc418/est3b4n/AQ_high_zpsf8361870.jpg) (http://s1210.photobucket.com/user/est3b4n/media/AQ_high_zpsf8361870.jpg.html)
(http://i1210.photobucket.com/albums/cc418/est3b4n/AQ_sun_zpsa845ecbc.jpg) (http://s1210.photobucket.com/user/est3b4n/media/AQ_sun_zpsa845ecbc.jpg.html)
Here it narrowly misses a swipe from a lightsabre.
(http://i1210.photobucket.com/albums/cc418/est3b4n/AQ_sun_2_zps64c7db41.jpg) (http://s1210.photobucket.com/user/est3b4n/media/AQ_sun_2_zps64c7db41.jpg.html)
We also got some footage from the on-board camera. It was somewhat shaky, kind of like an ongoing mirage effect, but otherwise it's OK to watch (if maybe a bit nauseating). Here's a still from when I think it was at around 50 feet or so during the first flight (the camera was accidentally attached upside-down, so I flipped the image). Balancing the individual rotor blades and adding some padding to the Velcro attachment might reduce some of the vibration.
(http://i1210.photobucket.com/albums/cc418/est3b4n/AQ_1st_flight_cam_zps7adbf139.jpg) (http://s1210.photobucket.com/user/est3b4n/media/AQ_1st_flight_cam_zps7adbf139.jpg.html)
We got about 3 flights out of it overall. I think they lasted for between about 2 to 4 minutes each but in theory it could probably manage about 10 minutes before a recharge.
The next thing we might try is to install the GPS receiver and see how well it handles "station-keeping" in the hover, and test out the barometer and ultrasound sensor for altitude holding. I've suggested that in order to make it easier to tell which way it's facing, the front two rotors could have some orange paint applied to them, and/or its shape could be made less symmetrical somehow. It already has port and starboard "nav lights" on the front two spars but in bright daylight they are not so easily seen, IMO.
-
OMG A BEAM! THEY ARE FIRING AT US! TAKE COVER!
-
OMNOMNOM FLYING SANDVICH HOLDER
-
Avoid the beam and you won't get hit, pilot!
-
@lostlama:
What software are you using for the quad?
Have you tried using rubber or plastic spacers/washers to mount the motor to reduce vibration? (Balancing the prop's would be a good start).
Those skids look VERY beefy compared to the rest of the quad, perfect for accidental hard-landings. :yes:
-
Well, that picture confirms it, someone's firing purple beams at our planet. :P I guess those guys didn't realize that their weapons were not finetuned enough to cause damage. ^^
That quadrotor looks pretty cool, by the way. Until the day we can deploy those as anti-drone measures, though, I'm not completely interested in building these or seeing much of 'em though. :P
-
Tupperware attack chopper!
-
That quadrotor looks pretty cool, by the way. Until the day we can deploy those as anti-drone measures, though, I'm not completely interested in building these or seeing much of 'em though. :P
Feeling up to the "Crowbar for every citizen" campaign? ;7
-
@lostlama:
What software are you using for the quad?
Have you tried using rubber or plastic spacers/washers to mount the motor to reduce vibration? (Balancing the prop's would be a good start).
Those skids look VERY beefy compared to the rest of the quad, perfect for accidental hard-landings. :yes:
This is actually my brother's project, not mine, but I've helped him out with a few tasks like calibrating the ESC, tethered flight testing, and taking camera footage. Building and flying R/C aircraft is one of his hobbies, and it's something I have a passing interest in. It uses the AeroQuad (http://aeroquad.com/content.php) flight software. I made a list of the IMU components it uses in an earlier post, if you're interested: http://www.hard-light.net/forums/index.php?topic=79831.msg1581164#msg1581164 (http://www.hard-light.net/forums/index.php?topic=79831.msg1581164#msg1581164)
I think there are only metal washers between the spars and the motors, so there's no real vibration-absorbing medium there. They're 290W each. The tupperware box itself is cushioned by some compressed silicon tubing.
The span that the original skids had was quite narrow (see the pic on the first page, they're just visible). These new ones are much more suitable.
Tupperware attack chopper!
Originally it was going to be one of those cylindrical packages for blank CD/DVD stacks but it wasn't the right size.
I doubt this is a hobby I could get into, certainly not now with my current level of tech knowledge, skills and lack of work space. If I were to get involved I'd probably be more interested in free flight modelling rather than R/C.
-
Besides the quadcopter-like layout, this isn't exactly drone related. But it's quite a feat IMO.
-
I want that. I'd bring a slightly modified weed-eater, of course... ;7
-
i want to know how they are controlling it. the obvious way would be to very the rotor pitch. but im not seing it.
-
The copter-cycle? It doesn't look like they're controlling it at all, it's just being moved by interior winds and Coriolis and what-have-you (probably not that).
-
apparently it did have some kind of a control system, the rotors had ailerons on the tips, though im not entirely sure how they are actuated. another interesting thing was the rotors were wound with some kind of exotic cordage, which were taken up by spools in the pedals. that means when you run out of string, you fall like a brick.
-
Drone-assisted dry cleaning delivery service (http://arbroath.blogspot.co.uk/2013/07/dry-cleaning-firm-uses-drone-to-deliver.html). I can't really see it being scaled-up, but it's an innovative advertising gimmick.
-
Just stumbled across this excellent demonstration of what is apparently the world's first quadcopter rescue (or should that be recovery?), showing some great use of a GoPro camera and FPV goggles.