A team of engineers at Google DeepMind has developed a robot that can play amateur table tennis. The core hardware of the robot is based on the ABB IRB1100 robotic arm, which is known for its high speed and high precision and is widely used in industrial scenarios. By combining advanced robotic technology and artificial intelligence, it is possible to perform fast and strategic table tennis movements.
By watching physical simulations and human competitions, training using small-scale data sets, and continuously improving skills through an iterative process. The robot easily defeated beginners when facing human players at different levels and won more than half of the games against intermediate players, but performed poorly against advanced players.
However, the competition seems to be less intense than the old man in the park.
For robots, table tennis requires the mastery of complex low-level skills and strategic gameplay, and requires long-term training. DeepMind believes that it may be a better choice to have a better strategy but be skillful in executing low-level skills. This distinguishes table tennis from purely strategic games such as chess and Go.
As such, table tennis is a valuable benchmark for improving robot capabilities, including high-speed movement, real-time accurate and strategic decision-making, system design, and direct competition with human opponents.
In this regard, Google DeepMind’s chief scientist praised: “Table tennis robots will help us solve high-speed control and sensing problems.”
If you want to learn more, you can click on the link below the video.
Thank you for watching this video. If you like it, please subscribe and like it. thank
Paper address:https://arxiv.org/pdf/2408.03906
Oil tubing:
