Robot learning has seen remarkable advancements in recent years; however, achieving human-level performance in terms of accuracy, speed, and adaptability remains a significant challenge across various domains. One such domain is table tennis—a sport that demands years of rigorous training for human players to reach an advanced level of proficiency.
In a new paper Achieving Human Level Competitive Robot Table Tennis, a Google DeepMind research team introduces the first robot agent that attains amateur human-level performance in competitive table tennis.
The team employed a hierarchical and modular policy architecture, which includes multiple low-level skill policies managed by a high-level controller. Each low-level skill policy is tailored to a specific aspect of table tennis, such as executing a forehand topspin, targeting with a backhand, or performing a forehand serve. These skills are built upon a shared foundational policy, allowing for further specialization as each skill improves.
The robot table tennis agent operates with a two-tiered control system, comprising a high-level controller (HLC) and several low-level controllers (LLCs). The LLCs are responsible for executing different table tennis skills by generating joint velocity commands at a frequency of 50Hz. The HLC’s role is to determine which LLC should be activated during each ball episode. Within the HLC, six components work together to decide the optimal LLC for the situation.
Beyond developing the policy itself, the researchers also collect and store data both offline and online about each low-level skill’s strengths, weaknesses, and limitations. These skill descriptors provide the robot with crucial insights into its capabilities and areas for improvement. The HLC uses this information, along with current game statistics and the opponent’s profile, to select the most appropriate skill at any given moment.
In their empirical study, the researchers evaluated the robot’s performance through 29 matches against human opponents. The robot won 45% of the matches (13 out of 29), including 100% of the games against beginners and 55% against intermediate players, showcasing a solid amateur human-level performance.
The paper Achieving Human Level Competitive Robot Table Tennis is on arXiv.
Author: Hecate He | Editor: Chain Zhang
We https://www.kdcleaningny.com/locations/ utilize the latest cleaning equipment and technologies to achieve the best results. From high-efficiency vacuum systems to advanced floor care machines, our tools enable us to clean more effectively and efficiently.
Robot learning has indeed made impressive strides in recent years. Despite these advancements, reaching head soccer human-level performance in accuracy, speed, and adaptability continues to present a significant challenge across different fields.
Despite its simplicity, dinosaur game still has some interesting features. As players reach high score milestones, the game scene will change from day to night, bringing a new feeling and adding motivation to overcome the challenge.
All parameters fluctuate strongly on the chart tap tap shots
No one can do better than you in drift boss game!
Delta Executor iOS is a roblox script executor for iPhones and iPads. It’s a special app that runs custom codes inside Roblox games. When you use it, you can do things that regular players can’t do. delta executor
The forge roblox calculator
is a customized tool for Roblox The Forge is called Roblox Calculator. For maximum crafting success, compute precise forging probabilities, forecast item stats, and optimize ore combinations.
The advancement of robot learning in table tennis is impressive, showcasing amateur human-level performance. Google DeepMind’s innovative approach is a major step forward in robotics. For more cutting-edge tech insights, visit steal brainrots and stay inspired!
I should mention how much I value this essay and how often I refer to it. You’ll probably keep posting better-than-average content here level devil.