Why March 11th Kuka will lose to Timo Boll
Hello.
Advertising campaign and KUKA marketing specialists have achieved their goal - many are discussing the upcoming match. Including people who do not represent the difference between ping-pong and modern table tennis, but are interested in modern technology.
In this mini-topic, I would like to try to explain why the upcoming event cannot be viewed as Kasparov vs deep blue. The first had at least some chances.
Arguments
0. This is an industrial robot created not for playing tennis. Deep Blue was originally designed to win in chess.
')
1. No one will write code that could compete with the tactics and strategy of a professional athlete. Yes, in table tennis there is a tactic, a strategy, and just a human cunning.
They will not write such a code, because it is unnecessary for marketers (and because it is expensive). To attract the attention of a wide mass of people to their products - a fairly well-made video and a world-class star.
( As offtopic, Timo Boll is one of the last hopes of Europeans against domination of China in this sport )
2. No one will design the technical environment of the robot:
- a set of full-hd cameras for image capture: there is not enough kinekta here - its resolution is not enough to detect small movements of the athlete's hand, which sets the angle and speed of the ball. If you make a mistake in analyzing this data, then in big-time sports, there will be no second chance.
- power for image analysis at different angles and decision making. This should be done extremely quickly => efficient glands, optimized code. Expensive to demonstrate the industrial robot.
3. The robot does not have an adequate platform for moving the manipulator. Even if we assume that the speed of the manipulator and its degree of freedom is comparable to similar human data, then the speed will not be enough to cover all the game situations / combinations. Legs are a support of the tennis player, base. In addition, it is the legs that expand the maneuverability of the athlete, give him the opportunity to have time there, where his hands do not yet have time.
( Pay attention to the muscle mass of tennis players' legs. This is no accident. )
Conclusion
I think these are the main points and it’s not worth waiting for the miracle on March 11th. The maximum that will be demonstrated is the ability to warm up, set some kind of ball rotation, defend against certain types of attacks and carry them out as effectively as possible (so that there is something to cut into a demo). Maybe even Timo, out of love for the art of robotics, will miss a couple of attacks.
But, objectively, if all of the above is demonstrated, it will be, without fools, cool. And allocate money, for example, in the same China, to develop the first version of the robot tennis player.
PS
Ping pong is not a modern table tennis.
Advertising campaign and KUKA marketing specialists have achieved their goal - many are discussing the upcoming match. Including people who do not represent the difference between ping-pong and modern table tennis, but are interested in modern technology.
In this mini-topic, I would like to try to explain why the upcoming event cannot be viewed as Kasparov vs deep blue. The first had at least some chances.
Arguments
0. This is an industrial robot created not for playing tennis. Deep Blue was originally designed to win in chess.
')
1. No one will write code that could compete with the tactics and strategy of a professional athlete. Yes, in table tennis there is a tactic, a strategy, and just a human cunning.
They will not write such a code, because it is unnecessary for marketers (and because it is expensive). To attract the attention of a wide mass of people to their products - a fairly well-made video and a world-class star.
( As offtopic, Timo Boll is one of the last hopes of Europeans against domination of China in this sport )
2. No one will design the technical environment of the robot:
- a set of full-hd cameras for image capture: there is not enough kinekta here - its resolution is not enough to detect small movements of the athlete's hand, which sets the angle and speed of the ball. If you make a mistake in analyzing this data, then in big-time sports, there will be no second chance.
- power for image analysis at different angles and decision making. This should be done extremely quickly => efficient glands, optimized code. Expensive to demonstrate the industrial robot.
3. The robot does not have an adequate platform for moving the manipulator. Even if we assume that the speed of the manipulator and its degree of freedom is comparable to similar human data, then the speed will not be enough to cover all the game situations / combinations. Legs are a support of the tennis player, base. In addition, it is the legs that expand the maneuverability of the athlete, give him the opportunity to have time there, where his hands do not yet have time.
( Pay attention to the muscle mass of tennis players' legs. This is no accident. )
Conclusion
I think these are the main points and it’s not worth waiting for the miracle on March 11th. The maximum that will be demonstrated is the ability to warm up, set some kind of ball rotation, defend against certain types of attacks and carry them out as effectively as possible (so that there is something to cut into a demo). Maybe even Timo, out of love for the art of robotics, will miss a couple of attacks.
But, objectively, if all of the above is demonstrated, it will be, without fools, cool. And allocate money, for example, in the same China, to develop the first version of the robot tennis player.
PS
Ping pong is not a modern table tennis.
Source: https://habr.com/ru/post/213221/
All Articles