Dimitrij Ovtcharov takes on incredible robot!

says Waldner Masterclass out now! 🏓
says Waldner Masterclass out now! 🏓
Well-Known Member
Administrator
Oct 2010
4,144
6,577
12,998
We have seen videos surface in the past of people taking on the impressive Omron Forpheus robot!

Well world number 3 Dimitrij Ovtcharov gave the robot a run for his money in this fascinating footage below!

<iframe src="https://www.facebook.com/plugins/video.php?href=https%3A%2F%2Fwww.facebook.com%2Ft2apac%2Fvideos%2F2047441778631888%2F&show_text=0&width=560" width="560" height="316" style="border:none;overflow:hidden" scrolling="no" frameborder="0" allowTransparency="true" allowFullScreen="true"></iframe>
 
This user has no status.
This user has no status.
Member
Dec 2016
496
388
1,723
This is the thread where we make fun of the robot, and someone digs this up within a few of years to make fun of us! ;)





There is more to it, though. Is it okay to go on a tangent?



Machines have always excelled at brute force. It's been a long time since anyone was last able to challenge a computer at chess (calculating power) and in physical terms, well, should I mention weightlifting?

But what happens when we can't program the thing to do any better, without investing enormous resources for marginal gains?
I assume most people are familiar with AlphaGo Zero's groundbreaking exploits, in a field where "in 20 years time" sounded like a conservative estimate for man-instructed software to catch up with the best players. So much for the "human" element...

Even when it comes to well-controlled environments - still using chess as an example - if any doubt remained as to what a couple of decades spent fine-tuning algorythms to absurd levels is worth against AI, this probably settles it (I won't get into the whole hardware debate, regarding that match; it makes no difference, at least in the longer-term).



So my not-so-concealed point is this: I assume we have the tools to make this robot fast enough to reach anything thrown at it. Sensors will improve, and so on with regards to all things hardware. But I can't imagine how in a million years we could instruct the thing to read spin, among other things, by feeding it lines of code.

... Can we get two of these to play against each other? By that I mean both robots having access to every little piece of data / input / output for either of them. Nicely controlled. Robot-1 hits the ball from position x, at speed y and in direction z, with a 90° bat angle. Robot-2 receives in a similar controlled way, and observes the result. Robot-1 hits the ball from position x, at speed y and in direction z, with a 89° bat angle. Robot-2 receives in the same way as the first time, and observes the result. Rinse and repeat till the end of times. Of course this dumb process can be optimised, which is also what humans do; but you get the point. Not so long ago I would have suggested a hybrid using what that fella in the posted video already knows, but I'm not convinced anymore.

Soon enough the thing will have to adjust to the non-linear behaviour of the ball, when compared to the stroke; it may pick up on its unexpected behaviour upon impact on the table; it may even learn to differentiate and adapt to new equipment, using a pre-determined set of strokes; or none of the above. I don't really have a clue, and that's beside the point. It's going to be clumsy, but what it learns, it never forgets and never fails to reproduce. Not to mention how that entire knowledge can then be replicated in the blink of an eye.

It's maybe unrealistic just now, and it's absurd if we only consider table tennis as a purpose. But the day machine-learning convincingly applies to the "real", physical world is when a whole lot of futuristic / borderline fantasy stuff from the time the steam-engine was invented to T-800, will come into shape. To me, it feels like the hardest part has already been done!
 
This user has no status.
This user has no status.
Member
Aug 2017
184
231
962
Five years until Dima can't make that robot miss. Fifteen years until the Butterfly Terminator prototype defeats the recently retired FZD in an exhibition match. Several months later it crushes the current world champion and ends the era of human dominance in table tennis.
 
This user has no status.
This user has no status.
New Member
Mar 2018
2
1
3
So my not-so-concealed point is this: I assume we have the tools to make this robot fast enough to reach anything thrown at it. Sensors will improve, and so on with regards to all things hardware. But I can't imagine how in a million years we could instruct the thing to read spin, among other things, by feeding it lines of code.

There are already pretty decent algorithms for identifying whats on the image, is it really almost impossible to read spin from the ball? What about to start with a ball colored with black lines to make it easier?
 
  • Like
Reactions: Ioiettino

erm

This user has no status.

erm

This user has no status.
Member
Apr 2018
42
19
70
Read 1 reviews
if elements like spin, deception...will keep being eliminated from the sport...better robots will be much easier to program...
 
This user has no status.
This user has no status.
Member
Aug 2017
184
231
962
There are already pretty decent algorithms for identifying whats on the image, is it really almost impossible to read spin from the ball? What about to start with a ball colored with black lines to make it easier?

Robots will learn to read spin the way humans do (visual and audio information from racket/ball contact, ball trajectory through the air and off the table, visualization of markings on ball surface), only they'll do it with more sensitive sensors and faster processors. Also possible for them to use senses that humans don't have (e.g., radar) but not necessary. And their reflexes will be fast enough to read spin directly from racket-ball contact and adjust racket angle and swing trajectory accordingly, but again not necessary. Just like Chess and Go players, we're doomed to defeat. On the bright side, within two or three decades we'll be able to go to the club and rent a robot coach or practice partner who can play better than Ma Long. It should somewhat level the international playing field, because everyone will have CNT-quality practice partners. Still won't be able to get the real national rubber though.
 
This user has no status.
This user has no status.
Member
Dec 2016
496
388
1,723
Five years until Dima can't make that robot miss. Fifteen years until the Butterfly Terminator prototype defeats the recently retired FZD in an exhibition match. Several months later it crushes the current world champion and ends the era of human dominance in table tennis.

I'd bet on less than ten years, time will tell!
 
  • Like
Reactions: Andy44
This user has no status.
I honestly thought tracking the spin was a solved problem? They've been using the Hawkeye ball tracking software in outdoor tennis and cricket for years and years now - and not just for replays for TV, they trust it for umpiring at the highest level! And it can track spin bowling in cricket.

Ok sure it's not directly reading the spin, but the flight path. But the flight path is affected directly by the spin, if the software knows the speed and path of the ball I think it could work out bat angle without too much trouble.
 
  • Like
Reactions: pingpongpaddy
says Spin and more spin.
says Spin and more spin.
Well-Known Member
Super Moderator
Dec 2010
16,146
17,685
54,747
Read 11 reviews
I guess there is more to it than just reading the spin. Reading the axis of spin, the amount of spin, and adjusting the angle of the racket to compensate for the spin.

The other thing is, the robot is making direct contact and the machine is moving the racket forward. The racket is going very slowly. With that kind of contact, any heavy spin will mess the robot up unless the angle of the racket is perfect.

Will they be able to make a robot that reads all the details of the spin, adjusts the angle of the racket and makes tangential contact with the racket accelerating to a high enough speed. When the incoming ball has heavy spin tangential contact and racket speed help control the spin.

Even in a good block you are not just putting the racket out there to let the spin take control over you. There is racket motion that is subtle and about how you touch the ball that makes the block solid.

It will be interesting to see how things develop. But I am not sure a robot can demonstrate the touch to counter heavy spin as the spin keeps changing. It will be interesting to see if they can do that.

But I know this robot Dima is playing looks fun. But would not be as good to train with as one of the high and Butterfly Robots.
 
Last edited:
This user has no status.
This user has no status.
Well-Known Member
Feb 2012
2,010
1,441
4,714
Read 1 reviews
lol table tennis is not chess or go. u simply play wide angles and its over. the money and time to build a robot that can play world class is not worth it and I highly doubt that robots can actually read or feel spin like a human does in the next 30-40 years. Even if programmers make it there is also the factor of tactics . Chess for example has limited best moves so its easier to calculate, in TT its a whole different story. different speeds spins tempos serves against each player . not so easy to adapt to so many variations
 
  • Like
Reactions: UpSideDownCarl
This user has no status.
This user has no status.
Member
Aug 2017
184
231
962
lol table tennis is not chess or go. u simply play wide angles and its over. the money and time to build a robot that can play world class is not worth it and I highly doubt that robots can actually read or feel spin like a human does in the next 30-40 years. Even if programmers make it there is also the factor of tactics . Chess for example has limited best moves so its easier to calculate, in TT its a whole different story. different speeds spins tempos serves against each player . not so easy to adapt to so many variations

You're right, table tennis isn't chess or go. Computationally in certain ways it's simpler because we know the relevant equations of motion and computers can easily solve them. This lets us design missile defense systems which perform a task often compared to hitting a bullet with a bullet. Much harder than hitting a ball with a racket. I'd agree we're still years away from the Butterfly Terminator, but not as many as you think.


 
  • Like
Reactions: JST
This user has no status.
This user has no status.
Member
May 2018
5
4
9
With advances in machine learning and computer vision, I wonder if in 10 years they will make a robot that could actually beat a pro. Lots of things seemed impossible in the past.
 
This user has no status.
This user has no status.
Member
May 2018
45
71
142
You're right, table tennis isn't chess or go. Computationally in certain ways it's simpler because we know the relevant equations of motion and computers can easily solve them. This lets us design missile defense systems which perform a task often compared to hitting a bullet with a bullet. Much harder than hitting a ball with a racket. I'd agree we're still years away from the Butterfly Terminator, but not as many as you think.

Second one video is great, very interesting for me!
 
Top