[HN Gopher] How can this 6 axis robot have a static accuracy of ...
___________________________________________________________________
How can this 6 axis robot have a static accuracy of 0.05 mm? (2021)
[video]
Author : unclefuzzy
Score : 248 points
Date : 2024-10-26 00:20 UTC (22 hours ago)
(HTM) web link (www.youtube.com)
(TXT) w3m dump (www.youtube.com)
| convolvatron wrote:
| that is a really great encoder trick. I wouldn't have thought it
| was that good, but it clearly is.
| Animats wrote:
| What their video demonstrates is mostly same-direction
| repeatability, not absolute static accuracy. They can correct
| for with backlash at individual motors, but not slop or bend in
| the linkages.
|
| This uses DC motors. If you use modern 3-phase servomotors, you
| know more of what the motor is doing.
| smolder wrote:
| I'm not a fan of the youtube link trend on HN, as cool as the
| latest robots are. I know they're encroaching on territory
| previously held by much heavier additive and subtractive
| machines.
| bob112 wrote:
| Are you saying you don't like video and would prefer text, or
| is it something specific to YouTube that you object to? For
| many topics, video is really helpful in understanding stuff.
| smolder wrote:
| It's a mix of both, I guess. I don't like youtube taking over
| for text based sources. It's less accessible, way less
| efficient, and feeds into the google surveillance machine.
| blamazon wrote:
| This doesn't address your (valid) systemic concerns, but on
| efficiency the way I like to use links like this is by
| adding them to a queue for later, then, if I have an
| opportunity to play them while doing the dishes or
| commuting or just feeling like watching a television I play
| them back, usually at higher speed of playback than 1.0. If
| I'm just sitting on the couch actively watching I'll read
| ahead in the transcript and skip ahead if I feel like it.
| d0mine wrote:
| Unlike the web, youtube has a functional search ;)
| defanor wrote:
| And I am okay with YouTube when a video makes sense, but in
| this case they have basically crammed a short article into a
| video, making it more awkward to read: slides with texts and
| diagrams, with some background music, and only a video
| demonstration in the end.
| avodonosov wrote:
| I also caught myself thinking that most of the content would
| be more accesdible as an article - I needed to pause and
| rewind several times. Although the aticle would include some
| video fragments (the final demo and some others)
| dr_dshiv wrote:
| I love hackernews
| franciscop wrote:
| I've seen in the past a different trick that is adding an IMU[1]
| to the robot arm. When combining two different types of sensors,
| it's called Sensor Fusion[2], and it's really common to put
| together a IMU with GPS and slap a Kalman Filter[3] for very
| accurate position reading.
|
| The particularly cool thing of this video though is that they
| could mount the new sensor within the motor itself, making it all
| a lot more compact.
|
| [1] https://en.wikipedia.org/wiki/Inertial_measurement_unit
|
| [2] https://en.wikipedia.org/wiki/Sensor_fusion
|
| [3] https://en.wikipedia.org/wiki/Kalman_filter
| shellfishgene wrote:
| He goes into detail on the control algorithm of the project on
| the github. It's rather complicated...
| https://github.com/adamb314/ServoProject/blob/main/Doc/Theor...
| acyou wrote:
| Great, but that robot isn't doing an actual task?
|
| I sort of struggle to see how getting good positioning accuracy
| from a high backlash system under zero load can have a useful
| application.
|
| Maybe just lack of imagination on my part.
|
| There is this trend that says make and buy bad hardware, the
| software will solve it. I haven't noticed that paying off. Tesla
| using webcams for self driving is an example. Boeing designing
| their planes and then using faulty attitude sensors is another.
|
| I would be way more impressed if the robot did something useful.
| My suspicion is that its real world application capabilities are
| rather limited.
| boeinggggg wrote:
| You have oversimplified the Boeing one: their goal was to
| create an efficient plane to compete with Airbus without
| needing the expense and delays of a new type certification.
|
| To do this they needed bigger engines on the same frame, which
| in turn needed to be mounted further forward affecting flight
| characteristics and requiring retraining. Retraining would be a
| sales killer so they hacked on some software systems to attempt
| to make the plane fly like an older 737.
|
| Then they can just use an iPad training course for pilots to
| upgrade. The augmentation had to avoid the pilot knowing about
| (I think) the plane getting stuck in a stall at a too high AoA
| (this is where my memory might be off...) so the MCAS software
| uses AoA sensors to nose down based on the detected AoA.
|
| The AoA sensors were never designed to be used for a direct
| life and death critical use case and sometimes they got stuck
| or failed. MCAS only used one as an input. If MCAS incorrectly
| asseses a nose down is required and the pilot follows their 737
| training they are having their last day. That plane is going
| down.
|
| Bascially people were murdered by Boeing so at every stage of
| this wretched plan they can make more money.
|
| I think you are right but Boeing was more of perhaps the worst
| possible asshole design, and deserves it's own league.
| zmgsabst wrote:
| There's some really negligent stuff, like changing how to
| disable auto pilot (ie, MCAS) -- as the pilots of both
| crashed planes attempted actions that would have disabled the
| autopilot on previous models.
| boeinggggg wrote:
| If the pilots know how this sausage is made, it aint a 737
| anymore. I think thay is the reason they rolled the dice
| sadly.
| vlovich123 wrote:
| Wasn't the Boeing issue completely preventable with an
| inconsequential extra part that cost nothing? Like the short
| cuts actually worked but they literally went all the way to
| almost succeeding and snatched defeat from the jaws of
| victory. (Aside from all the other things they did that also
| contributed to disaster situations going worse)
| boeinggggg wrote:
| I don't know. Maybe an expert can chime in but I think it
| is a hard problem because of ice etc. I think the 737Max
| has the problem where AoA matters more because you can get
| into a stall you can't get out of.
|
| Whereas maybe before on older planes you get in a stall and
| you nose down to reduce AoA. You don't need a sensor to
| know this look at altitude etc.
|
| So now you need perfect ten nines of reliability AoA
| sensors. Their use case has gone from a data point to
| mission critical, but the sensor is the same.
| imoverclocked wrote:
| You never want to get into a stall in a large commercial
| jet. Private pilots are taught stall and maybe spin
| recovery techniques for small GA aircraft. ATP rated
| pilots are taught stall/spin avoidance.
|
| Chances are, if your AoA is anywhere near the critical
| AoA, a competent pilot is likely aware of it. The sensors
| are just another safety factor on top of that to help
| ensure situational awareness.
| K0balt wrote:
| Or, in the case of the 737Max, to trigger a chain of
| events that proved lethal to hundreds to people. That's
| the secondary use of the AOA sensor in combination with
| the FC software that they implemented. It would have been
| relatively easy to integrate the AOA input with other
| sensors to eliminate this problem, but it would have
| invited a deeper look at the hazards of their design
| decisions.
|
| Bean counters bathing in blood, all the way down.
| robocat wrote:
| > Bean counters bathing in blood, all the way down.
|
| No resource is infinite and money is an important
| constraint in any engineering project. Engineering is all
| about making compromises. Good engineering is making the
| right compromises: especially when life and death
| decisions are being made.
|
| Casually blaming "bean counters" is a distracting fantasy
| available to anyone that doesn't have to make real-world
| decisions. Understanding the causes of how Boeing
| systematically screwed up requires a bit more maturity
| than you appear to show. "Bean-counters" particularly
| comes across as childish name-calling to me, and cliches
| don't help either.
|
| The fact that the MAX has been cleared to fly again shows
| that the design decisions were not utterly flawed.
| K0balt wrote:
| The design decisions were acceptable, if they had
| admitted the fact that the new design necessitated
| significant new training for the pilots, who were now
| flying a version of the 737 that could lose positive
| stability in some corners of its flight envelope....a
| fact they buried to reduce scrutiny (or facilitate
| deniability) from regulators and to make it an easier
| sell to airlines.
|
| Bean counters bathing in blood, all the way down.
|
| The forward mounting of the engine nacelles could have
| been countered with a small adjustment of the sweep or
| the surface area of the horizontal stabiliser, instead of
| the faulty flight control software solution, keeping the
| aircraft an aerodynamically safe aircraft as had been
| earlier generations. But that would have been a de-facto
| admission that the fundamental aerodynamic
| characteristics of the aircraft as certified were changed
| by the forward mounted nacelles.
|
| They chose to monkeypatch the flight control system
| instead of making a minor change that would have produced
| the inherently safe aerodynamic characteristics that the
| aircraft was certified with.
|
| They did this to avoid the delay and cost that would have
| resulted if they had been required to prove the aircraft
| design was still airworthy. There's a reason that new
| designs must be certified to be used in passenger
| transport. They tried to work around the fact that the
| 737 max is a substantially new aircraft by monkeypatching
| the FCS to compensate for a potentially dangerous
| aerodynamic flaw that was introduced by the new location
| of the engines.
|
| They chose to produce a more profitable but potentially
| dangerous aircraft instead of letting the engineers do
| their job and make the aircraft stable with the new
| engines. Regulators were also complicit in the regulatory
| evasion. Hundreds died as a direct result of this
| malfeasance.
|
| Bean counters bathing in blood, all the way down.
| sokoloff wrote:
| > If MCAS incorrectly asseses a nose down is required and the
| pilot follows their 737 training they are having their last
| day. That plane is going down.
|
| Boeing's argument is that an MCAS trim runaway is able to be
| addressed by the (memory item) Trim Runaway checklist and the
| crew of ET302 correctly used the STAB TRIM CUTOUT on that
| checklist during their attempt to save the flight. They then
| undid that action, in order to manually command nose-up trim
| (also reasonable under the circumstances, though contrary to
| the checklist), then stopped commanding nose-up trim while
| leaving the trim runaway checklist item reverted, allowing
| MCAS to continue the trim runaway that they'd previously
| correctly stopped by following basic 737 training. Then the
| flight was lost.
|
| Boeing did wrong here, but their argument was that if a 737
| pilot correctly executed the emergency checklist that is
| drilled into them during initial type training and in
| recurrent training, they'd be able to overcome that
| emergency. That falls into at least the probably technically
| correct category to me.
|
| (The yoke displacement method to disconnect the autopilot was
| not part of the emergency checklist for stab trim runaway.)
| pas wrote:
| Arguably the problem is that Boeing absolutely and utterly
| failed to do what they set out to do. After all, if the
| MCAS failures would present like the usual 737 runaway
| stabilizer, then the certified pilots would have been able
| to handle it as such. Since the "runaway MCAS" was a
| completely new phenomenon (one factor being the absolutely
| idiotic "on for a few seconds and then off for some"
| cycle).
|
| And as we know the FAA also was clueless, as they approved
| Boeing's "safety analysis".
|
| >>> Extensive interviews with people involved with the
| program, and a review of proprietary documents, show how
| Boeing originally designed MCAS as a simple solution with a
| narrow scope, then altered it late in the plane's
| development to expand its power and purpose. Still, a
| safety-analysis led by Boeing concluded there would be
| little risk in the event of an MCAS failure -- in part
| because of an FAA-approved assumption that pilots would
| respond to an unexpected activation in a mere three
| seconds.
|
| And, just to drive whatever point home, on top of all this
| the FAA completely dropped the ball, because it did not
| notice that they allowed Boeing to break their own base
| conditions which in effect invalidated the safety analysis.
|
| >>> As Boeing and the FAA advanced the 737 MAX toward
| production, they limited the scrutiny and testing of the
| MCAS design. Then they agreed not to inform pilots about
| MCAS in manuals, even though Boeing's safety analysis
| expected pilots to be the primary backstop in the event the
| system went haywire.
|
| It's understandable that Boeing wanted to avoid simulator
| training, but apparently this regulatory discontinuity (ie.
| either same or different, no in-between, as far as I
| understand) forced them to concentrate so much on avoiding
| the need for new type certification that they ended up
| completely believing their own crazy tale about the two
| models' sameness, which led to hiding information from
| pilots.
|
| https://www.seattletimes.com/seattle-news/times-
| watchdog/the...
| sokoloff wrote:
| I think it may have been a contractual term where Boeing
| could avoid a $1M reduction in purchase price per
| aircraft (times 280 aircraft) if simulator training could
| be avoided for the launch customer, Southwest Airlines.
|
| https://www.sciencedirect.com/science/article/abs/pii/S10
| 575...
| shellfishgene wrote:
| Look at this later videos where he has the servo lift a weight
| on a long arm.
| avodonosov wrote:
| Eurika! Other videos! Thank you for the idea.
|
| One example of real work:
| https://www.youtube.com/watch?v=GCHXNcpq3OA
| leovailati wrote:
| Regarding applications for robots that have to move very
| precisely without carrying a load, there are robotic
| measurement systems: https://en.wikipedia.org/wiki/Coordinate-
| measuring_machine
| gaze wrote:
| You can't judge backlash by how the robot repeats the exact same
| set of movements over and over. That removes hysteresis from the
| problem definitionally.
| shellfishgene wrote:
| But they're not the same motions? The second move is to the
| other side.
| KeplerBoy wrote:
| How do 3 axis robots you can buy for 100$ (3D printers) have a
| static accuracy of 0.05mm?
|
| It's not control theory, but mechanics and steppers.
| mmoustafa wrote:
| It's the compromise between a gantry vs arm design
| KeplerBoy wrote:
| There are plenty of 3d printers without a complete gantry.
| The bambulab a1 mini or prusa mini to name just two.
| alpaca128 wrote:
| Those two printers are also smaller and not as accurate at
| high speeds. The A1 Mini's slicer automatically places
| parts close to the Z axis in an attempt to reduce the
| issues and it uses input shaping, but given the printer can
| lift itself off the ground at default speeds that's not a
| perfect solution either.
|
| There's a reason the larger and faster printers often use
| the CoreXY design instead.
| KeplerBoy wrote:
| Of course, there's always trade offs in (mechanical)
| design choices. But their static accuracy absolutely is
| that good, which is fascinating at that price point.
| foxglacier wrote:
| Yes but they're not robot arms so it's not as fascinating. The
| length of the arm amplifies error so if you made a "mechanics
| and steppers" arm with the same positional accuracy as a
| printer, the motors would have to be much more precise or if
| you geared them down, the backlash extremely low like an
| industrial robot arm.
| KeplerBoy wrote:
| Sure, there's no free lunch.
| maille wrote:
| I'm curious what would be the best way to replicate an x/y
| (optionally z) system with 0.05mm or lower accuracy? Without
| sacrificing speed of course.
| sadhorse wrote:
| Depends on how much power you need (speed times force,
| acceleration and deceleration...) and how much stiffness you
| need (can't bend?).
|
| Also depends on how much travel you need. It is easier to get
| 50 micron accuracy over a total length of 100 micron compared
| to a total length of 1 meter.
| KeplerBoy wrote:
| Making things lightweight is crucial, because otherwise
| inertia will make your toolhead deviate from the planned
| trajectory at any corner.
| imoverclocked wrote:
| Question for anyone who has used one of these analog measuring
| devices: the indicator seems to go all the way around before the
| camera zooms in to read the indicated value. Is this video
| actually showing the accuracy it is claiming?
| lazide wrote:
| I haven't watched the whole video, but I'm assuming what they
| were showing was 'move x from 0.00 to 10.00' with the gauge
| showing the final move was to (actual) 10.05.
|
| Which with how floppy that rig is, is pretty impressive.
|
| Notably though, those gauges do need to be 'preloaded'
| (compressed into their 'positive' range) to be able to measure
| negative direction shifts, and while it looks like that was
| done, I can't be 100% sure without analyzing it far more than I
| want to do right now.
|
| Also, those gauges provide a degree of preload (not much, but
| some), which might be taking a bunch of slop out of the system
| and giving overly rosy accuracy numbers.
| toxik wrote:
| I think it's okay that they use the contract force to remove
| backlash since they are actually controlling it. In fact, if
| you could do that well, that's huge!
| lazide wrote:
| I don't think they could do that sustainably while it's
| actually doing 'the job' though, correct? It's pretty in
| the way.
| llm_trw wrote:
| If anyone wants to build this sort of thing the new Raspberry Pie
| Pico 2 is both orders of magnitude more capable than the chip
| used here and also around half the price.
|
| It's by far the best value for money for an introductory 32bit
| ARM/Risk embedded device right now.
| chipdart wrote:
| > (...) the new Raspberry Pie Pico 2 is both orders of
| magnitude more capable than the chip used here and also around
| half the price.
|
| That's cool and all but what are the tradeoffs?
| moffkalast wrote:
| Well it's new, it'll take a while for support to stabilize.
| The Pico 1 took almost a year given that the whole concept of
| a PIO was new, but this one should get there far sooner.
| 5ADBEEF wrote:
| Be aware, per erratum 9, that you'll need to include external
| pull-downs instead of using the internal ones
| nativeit wrote:
| It's relatively old at this point, but I'm still getting
| excellent performance from the Teensy 4.1. It's a little more
| expensive, around $30, but runs Cortex M7 @ 600MHz and includes
| a generous compliment of I/O and protocols.
| guenthert wrote:
| I don't see how the second sensor would improve accuracy (or
| rather precision). iiuc, the second sensor allows for improved
| speed. Backlash of the motor (and gears and linkage) could be
| accounted for using a PID controller, no?
|
| Said that, I'm impressed how precise this rather flimsy looking
| robot actually is.
| enginoor wrote:
| There are larger industrial robots that use secondary encoders to
| improve "out of the box" accuracy for more demanding tasks. The
| secondary joint feedback is paired with a kinematic model of the
| robot structure/mechanics to accurately predict where the robot
| tool point actually is.
|
| https://electroimpact.com/Products/Robots/AchievingAccuracy
| amelius wrote:
| Of what use are the primary encoders then?
| xattt wrote:
| If I'm not mistaken, one encoder measures the position and
| force applied by the motor, while the other encoder measure
| the position of the slack of the business end of the robot.
| amelius wrote:
| Doesn't this youtube project infringe on the patent which
| this company holds, then?
|
| https://electroimpact.com/Company/PatentFiles/US8989898B2.p
| d...
| sadhorse wrote:
| Legit question: if I replicate a patent for a personal
| non profit use, is this infringement? Perhaps it is
| because I'm benefiting from the intelectual property.
| amelius wrote:
| !> Yes, replicating a patented invention, even for
| personal, non-profit use, is technically considered
| patent infringement. A patent grants the inventor the
| exclusive right to make, use, sell, and distribute the
| patented invention for a certain period (usually 20 years
| from the filing date).
| Luc wrote:
| Depends where you live. E.g. in France there's a personal
| use exemption, in the US there mostly isn't.
| RobotToaster wrote:
| In the USA there's an exemption for research use of
| patents, specifically for "amusement, to satisfy idle
| curiosity, or for strictly philosophical inquiry."
| https://en.wikipedia.org/wiki/Research_exemption
| metalman wrote:
| Call it what it is,investigative journalism
| WJW wrote:
| How is that patent even a thing from 2009? Position
| feedback in robots is WAY older than that. I have
| textbooks at least a decade older than that patent
| describing that very system, so I wouldn't be surprised
| if it falls over at the first prior use claim it
| encounters.
| EasyMark wrote:
| How in the world can a company get a patent on math and
| basic techniques that have been around for decades before
| the patent was even filed? I can understand materials,
| unique "first come" algos, brand new mechanics, but there
| nothing novel in that patent. There's nothing novel about
| having secondary (or tertiary or ....)feedback for a
| system
| Karliss wrote:
| What part of the patent in your opinion infringed by the
| youtube video?
|
| Robot arms have existed long before 2015. And a lot of
| them use some combination of encoders. The term
| "secondary feedback" by itself without clarification
| doesn't really mean anything specific, and in terms used
| by the patent I would call this more of adding
| primary/primary feedback system. The part that the patent
| seems to repeat is having secondary position sensor
| attached to the mechanical joint of robot (I assume as
| opposed to encoders already builtin into the servo
| drive), although patentability of even that seems
| somewhat questionable in 2015. I am not that good at
| reading patents, so maybe I am missing the actually
| relevant/novel part of that patent.
|
| In the video both encoders are builtin the servo, instead
| of attached to the arm itself, even more the extra angle
| sensor introduced by author is attached directly to motor
| before the gearbox and it's slop which is complete
| opposite of what the patent tries to claim. The angle
| servo attached to output shaft after gearbox is what all
| hobby servos have.
|
| If you go through the actual claims of patent most of
| them are not applicable to the video. 1) "system for
| large-scale assembly operations, ... secondary feedback
| mounted to joint ...". Not suitable for large scale
| asembly operations, no feedback attach to joint, both
| feedback systems are built in the hacked servos and can't
| measure any slop within the joint itself or servo to
| joint connection. 2) angular accuracy of 0.05 arcminutes
| - very unlikely 4) system of claim 1 wherein the
| manufacturing assembly is an aerospace assembly - no
| aerospace assembly making here, 5) 6rotary axis and 1
| linear axis - no linear axis, 6) secondary feedback
| system is optical encoder -> questionable whether the
| optical angle sensor attached before gearbox matches the
| definition of "secondary encoder" as described by rest of
| the patent, also optical encoders is typically used for
| describing relative postion/angle sensor based on bunch
| of slits and counting pulses instead of analog amplitude
| measurement which gives absolute position. Typically I
| wouldn't bother with minute differences in classification
| of how the angle sensor is implemented, but since patent
| explicitly lists very specific sensor technologies I
| guess it matters. Otherwise they could just claim that
| there is an angle sensor/encoder. 7) secondary feedback
| system is inductive encoder - no inductive encoders here,
| 8) magnetic encoder - no magnetic encoders, 9) secondary
| feedback system is resolver - no resolver here (as in
| analog angle sensor based on ac coupling change depending
| on angle between two parts to directly generate the
| sin/cos of angle), 10) "system for acurate large scale-
| manufacturing assembly operations, ... >3 axis robot arm,
| with end tool, secondary feedback mounted on rotary
| joint" - this just more or less restates claim 1 only
| this time mentioning >=3 instead of >= 6 axis for some
| reason and mentions an end tool. Is ballpoint pen an end
| tool for large-scale manufacturing operations? Also the
| secondary feedback thing discussed before.
| ipsod wrote:
| It lets you see the position of the motor's shaft. That's
| used in some motor control algorithms, even if the motor's
| position isn't exactly the joint's position.
| luikore wrote:
| With direct drive (no gear box), do we still need the secondary
| encoder?
| mglz wrote:
| No, but you will almost certainly not get the required torque
| for a robot arm.
| TacticalCoder wrote:
| I've always wondered... Why aren't pantographs used more with
| robots when precision is needed?
|
| It's used to cut precise wood pieces or carve wood or metal etc.
|
| https://en.wikipedia.org/wiki/Pantograph
|
| https://youtu.be/s56J_Rnh_Co
|
| You use the "big" part to drive the "small" one, which gives it
| great precision.
| abecedarius wrote:
| Feynman brings up some problems with pantographs and precision
| in https://calteches.library.caltech.edu/1976/1/1960Bottom.pdf
| (page 6-7). I haven't thought about it myself, though.
| whamlastxmas wrote:
| Pantographs are mostly useful in 2 dimensions and any robot
| only needing two dimensions can just use rails which will be
| more accurate
| elif wrote:
| I have a hunch that Optimus likewise leans heavily upon inverse
| kinematic modeling. However not using the paper plate tech.
|
| It would be sick if they use a pure vision ML approach to train a
| heuristic understanding of its own muscles, instead of these
| fixed rotary encoders which do not account for material
| deflection, sensor dislodgement, etc. sort of like meta quest
| player tracking in the SLAM loop.
| iancmceachern wrote:
| In addition to this technique you can also use kinematic
| calibration which takes it to a whole other level.
| dreamcompiler wrote:
| Closed loop feedback is the key to high robot accuracy with cheap
| parts. The big trick is the position detector, and this person
| figured that out.
|
| Kudos!
___________________________________________________________________
(page generated 2024-10-26 23:01 UTC)