• Welcome to the Speedsolving.com, home of the web's largest puzzle community!
    You are currently viewing our forum as a guest which gives you limited access to join discussions and access our other features.

    Registration is fast, simple and absolutely free so please, join our community of 40,000+ people from around the world today!

    If you are already a member, simply login to hide this message and begin participating in the community!

Robot scramblers in future WCA competitions

abunickabhi

Member
Joined
Jan 9, 2014
Messages
6,713
Location
Yo
WCA
2013GHOD01
YouTube
Visit Channel
If say in a few years, we are able to make accurate scrambling robots, that can take input from TNoodle, and perform the scramble on any given cube for a round, what should be the ethics or rules that we should be formulating?

The scrambling bot should be able to scramble all types of puzzles (stickerless/black, 55mm/57mm, 3x3/sq-1), and it may have also some OCR tech to recognise which scramble has to be performed for that cube, and say a stamp to attest at the scrambler’s signature in the scorecard.

In the past, many solves have been misscrambled and some of them have even been WRs like Felik’s OH solve or Max’s OH solve. And also to scramble larger puzzles like 7x7, having human scramblers slows down the competition a lot.

There are lot of technological and logistic challenges to overcome before scrambling bots become a reality, but we should have some idea on how to adopt it into WCA competitions seamlessly. The main issue will be producing these bots at scale to deploy at every competition which will be a major problem, and also the cost of developing it. Fortunately, many cubers including myself are techies so we might develop it at bare minimum cost.

As a community, what kind of issues or situations, do you think will arise, once scramblers role is no longer a part of volunteering, and done independently by some machine.

More discussion on this topic on: https://forum.worldcubeassociation.org/t/robot-scramblers-in-future-wca-competitions/2409
 

CodingCuber

Member
Joined
Nov 10, 2019
Messages
703
Location
Melbourne
WCA
2019HERR14
If say in a few years, we are able to make accurate scrambling robots, that can take input from TNoodle, and perform the scramble on any given cube for a round, what should be the ethics or rules that we should be formulating?

The scrambling bot should be able to scramble all types of puzzles (stickerless/black, 55mm/57mm, 3x3/sq-1), and it may have also some OCR tech to recognise which scramble has to be performed for that cube, and say a stamp to attest at the scrambler’s signature in the scorecard.

In the past, many solves have been misscrambled and some of them have even been WRs like Felik’s OH solve or Max’s OH solve. And also to scramble larger puzzles like 7x7, having human scramblers slows down the competition a lot.

There are lot of technological and logistic challenges to overcome before scrambling bots become a reality, but we should have some idea on how to adopt it into WCA competitions seamlessly. The main issue will be producing these bots at scale to deploy at every competition which will be a major problem, and also the cost of developing it. Fortunately, many cubers including myself are techies so we might develop it at bare minimum cost.

As a community, what kind of issues or situations, do you think will arise, once scramblers role is no longer a part of volunteering, and done independently by some machine.

More discussion on this topic on: https://forum.worldcubeassociation.org/t/robot-scramblers-in-future-wca-competitions/2409
As I replied to this topic,

I like the idea of this. One problem I have is how easy it would be for people to connect to it. It would be most practical for it to connect to TNoodle via bluetooth (many already exist ie GAN robot, etc), however this is something that can easily be interfered with. Bluetooth devices can be detected by any smart phone, watch or computer and easily connected to. It honestly seems like a lot of extra work. An alternative solution would be a simple cube scanner that can check after a cube has been scrambled if it is the correct scramble (although implementing this with non-regular colour schemes would be difficult)

btw, I don't think it is necessary to move all these WCA Forum posts to here. The WCA Forum is specifically for posts about the WCA and competitions, not speedsolving in general.
 

abunickabhi

Member
Joined
Jan 9, 2014
Messages
6,713
Location
Yo
WCA
2013GHOD01
YouTube
Visit Channel
As I replied to this topic,

I like the idea of this. One problem I have is how easy it would be for people to connect to it. It would be most practical for it to connect to TNoodle via bluetooth (many already exist ie GAN robot, etc), however this is something that can easily be interfered with. Bluetooth devices can be detected by any smart phone, watch or computer and easily connected to. It honestly seems like a lot of extra work. An alternative solution would be a simple cube scanner that can check after a cube has been scrambled if it is the correct scramble (although implementing this with non-regular colour schemes would be difficult)

btw, I don't think it is necessary to move all these WCA Forum posts to here. The WCA Forum is specifically for posts about the WCA and competitions, not speedsolving in general.
I moved a few interesting posts here, as there are a few people who do not discuss on the WCA Forums, and I wanted to get the more ideas into these topics.

I will not be importing all the discussions from WCA Forums and WCA Github,but just the ones, I feel need to be discussed a bit more, or needs to be revisited.
 

qwr

Member
Joined
Jul 24, 2019
Messages
3,371
YouTube
Visit Channel
I cannot seeing it be practical in the next few years, just because all the cubing robots that have been built have been expensive one-off projects. The GAN robot is the only mainstream product and it's expensive. I believe a robot is just something that's too niche and expensive, compared to money better spent on venues and getting stackmat timers. Mainstream robot use is quite limited, with most robots being used in controlled industrial environments and performing work 24/7. The most mainstream robot is a roomba and that is much simpler to design than a cube scramble bot. And one robot is not enough per competition so comps would need at least several.
 

RoundUpCubing

Member
Joined
Jan 10, 2021
Messages
103
Location
Switzerland
YouTube
Visit Channel
I cannot seeing it be practical in the next few years, just because all the cubing robots that have been built have been expensive one-off projects. The GAN robot is the only mainstream product and it's expensive. I believe a robot is just something that's too niche and expensive, compared to money better spent on venues and getting stackmat timers. Mainstream robot use is quite limited, with most robots being used in controlled industrial environments and performing work 24/7. The most mainstream robot is a roomba and that is much simpler to design than a cube scramble bot. And one robot is not enough per competition so comps would need at least several.

With needing several robots, you might be thinking a bit too small. Nobody ever specified how many cube slots it should have, you could have one robot scramble multiple cubes at once. Financial/design aspect does come into play here tho.
 

abunickabhi

Member
Joined
Jan 9, 2014
Messages
6,713
Location
Yo
WCA
2013GHOD01
YouTube
Visit Channel
We now know that 2x2, 3x3, bigcubes and clock can be successfully moved/turned/scrambled.
I still do not know if a sq-1 scrambler exists. It would be super hard to build as the puzzle shapeshifts.







Same slot 2-5:
 

abunickabhi

Member
Joined
Jan 9, 2014
Messages
6,713
Location
Yo
WCA
2013GHOD01
YouTube
Visit Channel
To prove how slow human solving of DNFs and scrambling is, I made this video. Timing is benchmarked on 180 cubes.
I am working on a robot that would scramble N cubes into different states and orientations, quickly so that organizing MBLD at comps become less of a hassle.

Also people would procrastinate less if there is a scrambling robot at home to do more MBLD attempts in a day.

 

Cuber987

Member
Joined
Oct 20, 2022
Messages
622
Location
a
We now know that 2x2, 3x3, bigcubes and clock can be successfully moved/turned/scrambled.
I still do not know if a sq-1 scrambler exists. It would be super hard to build as the puzzle shapeshifts.







Same slot 2-5:
The thing is, these robots for pyraminx and megaminx (and big cubes) take way too long. This would make the comp take even longer if we scrambled like this.
 

OldSwiss

Member
Joined
Oct 10, 2022
Messages
361
Location
Switzerland
I think the biggest problem would be the different sizes and types of cubes. The GAN Robot can only scramble some 3x3 GAN Cubes, homemade Robots are usually slow and also only for one Cube Size.
And also the tournament organizers or the WCA had to spend a lot of money to buy some robots.

An other technical options to avoid misscrambles could be an app that checks the scrambled cubes by camera.
This way you would find bad scrambles quite fast but if it is missscrambled, you have to solve and rescramble it.
That could be a problem especially for bigger cubes, but you also have that problem with manual checks.
 
Joined
Aug 12, 2013
Messages
5,090
Location
Brazil
SS Competition Results
YouTube
Visit Channel
To prove how slow human solving of DNFs and scrambling is, I made this video. Timing is benchmarked on 180 cubes.
I am working on a robot that would scramble N cubes into different states and orientations, quickly so that organizing MBLD at comps become less of a hassle.

Also people would procrastinate less if there is a scrambling robot at home to do more MBLD attempts in a day.

hmmm a video was needed to prove that point?
 

erictom333

Member
Joined
Apr 11, 2023
Messages
92
Location
Not telling you
I think robot scramblers would make the most of an advantage for events with long scrambles with shallow-cut puzzles, most notably Megaminx and Curvy Copter. These have the advantage of being able to be scrambled by pronged arms that stay engaged at all times like the GAN robot, as opposed to arms that extend, turn, and then retract.
 
Top