My Safest "Un-Safe" Project

[@jiunuope]

Never imagined using it this way...

[@frommarkham424]

Wow

[@unspeakablerja8556]

How'd you get that much money?

[@DavesArmoury]

That's my favourite way to use something, outside of "manufacturer's recommended use" :)

[@jiunuope]

@@DavesArmouryTechnology's risks usually damage the weaker, so we need more manufacturers like you who focus on safety.

[@DavesArmoury]

​@@jiunuopeit's so easy to skip safety, but it can't be like that, especially with a system like that. Ended up splitting this project into multiple videos to make sure that I cover the important bits properly

[@Muschelbu]

You cover some basic safety issues, that’s great. But as a ride controls engineer I see several issues: How do you limit the accelerations and decelerations, as well as the max. speeds in a safe way to limit brake distances and keep g-forces in an acceptable range? For sure the lasers are not usefull for that job and the internal logic of the robots are not reaching the required performance levels at all. What do you do in case a holding brake fails? How do you prevent riders to crash against the robot arm itself? How do you prevent unintended start of the robot, also while not secured? Mostly depends on your starting mechanism and procedure. How do you limit the deceleration in case of a power failure? Just in case, the emergency stop is not a replacement for any safeguard that is required based on the risk assessment and hazard analysis. I see all these issues and more at least not covered by the video and potentially put riders into a serious safety hazard, if not properly dealt with it. You project is awesome, but safety has to be taken serious at all times.

[@jwrm22]

I don't have experience with robotics myself, but have designed functional safety for both automotive and industrial equipment. I agree with your assessment, it's safer, but not at all safe enough to be used like this. Many of the 'if this than that' in the video are in itself single point of failures and an inherent safety risk. I'm largely thinking of where the motion is stopped, but the stopping point is far from ideal. The author has commented on others that there isn't a need for a cage for the rider, as other safeties cover this. But the idea is to reduce the chance and the impact to within reasonable limits. This may also include a thick rubber matt under the rider, a solution to get the rider back in a safe position after a failure, and external monitoring which cuts off the robot, etc. But this likely means the project will never get off the ground.

[@DavesArmoury]

Agreed. There is a bunch of stuff that isn't included in the video, but is present. The safety lidars would ideally never be used. There are position, velocity, and torque limits on all the joints, and hard-stops were possible. A safeguard stop (not emergency) has a ramp down time to slow to a stop.

[@ExcessiveOverkill]

@@jwrm22I believe the commercial Kuka robots have externally accessible brakes and motor shafts so you can manually move the arm in case of a power failure. Not sure when it comes to the deceleration limits though, you’re probably just stuck with it stopping pretty hard, but if you’re not moving very fast then it might not be much of a risk.

[@ExcessiveOverkill]

I don’t know if Kuka has similar functionality to fanuc or how safety levels are classified for rides, but many newer robots do have SIL3 rated controllers with redundancy for making sure proper limits are maintained. Loss of power is a whole other issue, you’re probably stuck with the few G’s of deceleration no matter what.

[@Legendary5678X]

@@DavesArmoury what country is it in?

[@otodusobliquus3836]

Now I am certain you NEED a bigger lab! Can't wait to see where this project goes as well!

[@louie_ooeygui]

I love the safety focus in the video - I wish more makers took safety seriously🙂 Cool project!

[@DavesArmoury]

You and me both! Lots of fun stuff coming now that nonesense can be done safely 👍

[@Chris-oj7ro]

There is no other machine that you should focus on safety for more than one you are strapping yourself into!

[@Andre-qo5ek]

time to hook it up to a flight sim. and VR.

[@DavesArmoury]

Very possibly already in the works :)

[@zablok69]

In Legoland Billund, you could set up what the arm should do in sequence using a simple menu on the tablet before strapping in. lots of fun :)

[@DavesArmoury]

Always wanted to, but never actually ridden a KukaCoaster, until now!

[@PumpiPie]

Yeah, i remember doing the same ride in legoland.

[@Yellow_bud]

GLaDOS is just chilling in the background :3

[@fanBladeOne]

Ok. I see where you're going with this. You are a genius, Sir.

[@bda249]

I highly recomend to create a protective cage. So atleast it will not crush your head if something goes wrong.

[@DavesArmoury]

That's the whole idea behind all this safety. If something does go wrong, there is redundant safety everywhere so you don't need a cage

[@bl2575]

I love the solution of using safety lasers to prevent getting crushed against the wall and the ceiling. Dead simple and no complex software layer that could be buggy. You can still end up, stuck upside down near the ceiling, but you are alive :-).

[@DavesArmoury]

Right, but the system can't be run by the rider, so there is always someone there on the controls to get you down :)

[@MD80-cockpit]

I have seen full professional cnc machines that run perfectly. Everything seemed perfect. But the user of the machine had found out that if you activated the emergency stop at a certain point, the machine would make a run away at run full speed towards the end stop. It's about the holes in the cheese and an unfortunate coincidence, then the accident happens. I don't want to sit in it until it has clearance all the way around it.

[@crazybird199]

This looks like a really cool project! I feel like it may be a good idea to have leg restraints as well to keep them from swinging around unpredictably. I'm not expert, so there could be problems with this idea.

[@DavesArmoury]

Not a bad idea. Like some foot pegs to "stand" on. Lots of rollercoasters do have free-hanging legs

[@crazybird199]

@@DavesArmoury yeah, that’s the idea, I will say roller coasters usually have better clearance, though

[@Kriakye2435]

This is really cool!

[@Brandtechnik]

EPCOT had this years ago, called Sum of all Fears. Very fun to program your own coaster then ride it.

[@DavesArmoury]

Yeah, Kuka does have a line of entertainment rides with their arms. They are quite expensive though

[@cozmicraccoon9014]

I have an idea! Maybe make it so you have vr on the ride so it’s even more realistic!

[@alexdales6759]

Another great video, Dave. Can't wait to see what you do next!

[@Legendary5678X]

For some reson I so want to have a go on this

[@josebayona7739]

Awesome

[@jotchava]

Yesss!!! I remember talking with you about this and now it exists!!! Freakin awesome dude!

[@Legendary5678X]

What country is it in?

[@christophererickson2606]

Really appreciate your commitment to safety. I was hoping to hear about what you did related to the joint and workspace limits, as that would be the next big safety concern of mine. How did you account for multiple body sizes (taller riders/ wider riders?) Is it software based? etc.

[@DavesArmoury]

Joint limits and workspace monitoring were fudged at first to get them close, then verified with the largest person I know. They are fairly conservative.

[@Legendary5678X]

I acctuly think this is so cool also mind me asking what country is it in?

[@SafeDodecahedron]

"Looks Like It Would Kill You, Its A Cinnamon Roll"

[@xara7644]

Yes, futoroscope in france has a kuka robot arm based music attraction that works exactly like this. Like... exactly...

[@DavesArmoury]

Yeah, kuka sells a certified version of their arms for these applications that you can buy. Not cheap, but certified

[@stevengongg]

🔥🔥🔥

[@piotrfiurst9527]

So I've seen you made glados, so can you make weathley or one of the cores

[@WizzKidxKOx]

Kuka time!

[@MrPati0]

Now make this work together with a flight sim in VR!

[@SafeDodecahedron]

2:51 I Did Not Expect To See That lol

[@michits1118]

This would be amazing with a VR headset

[@DavesArmoury]

I do have plenty of plans for it 😀

[@randomrouting]

Yees! I gotta watch this again when i got sound but I LOVE IT ALREADY! Been thinking of doing something like this + a vr headseat, would be sick!! Thanks for sharing!

[@DavesArmoury]

I have a lot more plans for this arm, and VR is definitely one of them :)

[@randomrouting]

@@DavesArmoury Great! Already got the robot so will be following closely. :)

[@tomapc]

KR150? Nice, i did that without a chair just hanging, it was terrifying !

[@DavesArmoury]

It was hard enough to get on it with a harness! Yeah, 150 l130

[@TheDutchMagicTeacher]

So, you're adding a VR headset to this right? And then rollercoater tycoon will be the best game ever

[@DavesArmoury]

Working on the VR headset, but running into a lot of technical hurdles. Slow and steady...

[@TheDutchMagicTeacher]

@@DavesArmoury any specific hurdles? I have a bit of experience with unity and VR. Even if I can't help, a community post asking for help might help more then you think, you have 1000's of brains at your disposal, one of them is bound to be able to help

[@DavesArmoury]

@@TheDutchMagicTeacher mostly robot control issues.

[@TheDutchMagicTeacher]

@@DavesArmoury unfortunately, that is not my expertise, but like I said if you throw out a community post you might get some experts who can help you

[@Legendary5678X]

@@DavesArmoury what country is it in?

[@PumpiPie]

How mutch power do it need?

[@DavesArmoury]

600v, 3-phase, but less actual power than you would think

[@Jpcooldood]

sick

[@ThumbnailGrail]

More lasers is more better!

[@DavesArmoury]

#facts

[@Legendary5678X]

@@DavesArmoury what country is it in?

[@zoeyzhang9866]

Dope!

[@SprocketN]

I work with industrial robots and have always wanted to do this too. If I did I’d be sacked 🙁 My garage isn’t big enough for me to do what you’ve done. I’ll follow this project with a hint of jealousy 😀

[@DavesArmoury]

It' been a dream forever. Took me buying my own arm and renting space for it to make it happen

[@WeTrudgeOn]

Are you kidding me? Safety with one inch angle iron? I've been a fabricator for 35 years and I don't care what fusion said one inch angle iron IS NOT going to hold up top that kind of repeated stress. Is your neighbor a certified welder? If not how much experience does he have?

[@mubasshir]

IKR? It looks sooo unsafe. I would not trust that seat. And as you rightly pointed out, repeated loading will cause the assembly to fail and has to be taken into account

[@DavesArmoury]

Welds every day for his job. Yes, the framing is a bit light for long-term use, but it will be getting an overhaul for v2 anyway. Any suggestions for material. Somewhat limited by payload capacity

[@WeTrudgeOn]

@@DavesArmoury at least 1.5x1.5x3/16 that would be more likely to hold up. The thing is with smaller angle there is less area for weld.

[@mythzel898bet2]

Sir i want to ask if you can give me your wisdom about how to go through the major of mechatronics (or just engineering wisdom in general).

[@DavesArmoury]

Figure out how you learn. Everyone does it differently. Do what works for you

[@badtyprr]

The hole in one safety video ended me. 😂

[@tomallo99]

That's so cool... What's the model number of the arm?

[@DavesArmoury]

KR 150 L130

[@HMPirates]

What controller are you using, definitely not a C2, we can't hear it in the video?​@@DavesArmoury

[@DavesArmoury]

@@HMPirates yeah, it's a C2 in another room. Oldie but a goodie

[@bowieinc]

I’ll just say your work space shown is similar to what I have to tinker in. Give the opportunity to install a robotic arm in my workspace, I would choose this exact robot arm size. :) #IWantOne :)

[@DavesArmoury]

It's definitely too big for the space, but planning ahead

[@Prosecutor_justice]

No safety in Aperture lab

[@gigilsu]

make a wheatly robot!

[@officalfrogstudos]

Where's glados

[@acejones5260]

1:18 here in the bg

[@Legendary5678X]

Just how the heck

[@srilemobitelsrile8809]

It would be better to show whole safety process.

[@DavesArmoury]

Agreed, but it is youtube. Aiming for a middle-ground. Informative, but still watchable

[@gyeongchankim5423]

Not every thrill rides are classified as roller coaster and this is definitely not a roller coaster. Nevertheless, impressive.

[@DavesArmoury]

Agreed, but this is just the first step :)

[@tamillesteadman9389]

Where is Gladys?

[@blazeboi3305]

Looks like fun. 👍

[@MyraLevin]

Um your not alone I think

[@jdubdoubleu]

You would think they would at least have helmets 🤷

[@DavesArmoury]

You need a helmet if their head might hit something. This whole video is about making that not happen

[@Legendary5678X]

We need a colab 💀

[@purrzival]

I recently dug into how big industrial flight simulators work, and if you plan to directly translate the movement of a rollercoaster to the arm, i dont think it's going to be that realistic. What the industrial flight simulators do, is simulating the g forces that a pilot experiences, this video is great for explaining the concept: kzbin.info/www/bejne/ppe3lX16Z7SUoaMsi=IihgcS7mqI1wI1yj

[@DavesArmoury]

Agreed. When I get there (working on it) I'll be using acceleration and orientation.

[@ghostbombl8034]

Kids becomes a hammer if it malfunction. 😅

[@YouBeautiful-ff6of]

Port

[@hignaki]

Good video, but the music was way too loud compared to your voice. It's impossible to watch on open speakers without actively adjusting volume. Thanks.

[@DavesArmoury]

Sorry about that. Thanks for the feedback. Always working to improve the edits

[@Legendary5678X]

Just how the heck