Skild AI Develops Self-Play Training for Robots to Master Physical Skills
- Published
Skild AI has introduced a self-play training method that enables robots to learn physical skills autonomously, utilizing its S1 robotics foundation model. This innovative approach allows robots to engage in simulated soccer for over 140 years, enhancing their abilities without human demonstrations.
The significance of this development lies in its potential to enable robots to acquire complex skills across various environments, such as factories and homes, without the need for human guidance. By using Nvidia Isaac Sim for simulation, Skild's robots learned to score goals through competition against progressively advanced versions of themselves, showcasing a unique learning mechanism.
Looking ahead, the successful transfer of learned skills from simulation to a physical humanoid robot marks a pivotal step towards achieving physical artificial general intelligence (AGI). Skild's approach, which emphasizes self-play over traditional human-generated data, could redefine how robots are trained in the future. No further timeline was disclosed at the time of publication.
Original report
Skild AI trains robots to learn physical skills through self-play
RoboticsAndAutomationNews.com · Sam FrancisThis briefing is an independently written summary based on publicly available reporting and is provided for industry information and news discovery. The original report and source publication are credited and linked where applicable. RobotToday does not claim ownership of third-party source material.
Rights concerns? Contact [email protected] with the relevant URL and details. We will review the matter and take appropriate action where warranted.
More related news
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
-
11
ACIRS 2027 – Hong Kong, Jul 23–25, 2027 RobotToday · 2 days ago
- 12
Related Suppliers
Skild AI
United States
Pittsburgh AI company building the Skild Brain, a general-purpose foundation model for controlling any robot embodiment.
Shanghai Artificial Intelligence Laboratory 上海人工智能实验室
China
Shanghai-based national AI research institute founded 2020; develops the InternLM model family and the Intern-Robotics embodied-AI engine.
GigaAI 极佳科技
China
Beijing embodied-AI startup building world models (GigaWorld) and a robot brain (GigaBrain) that power humanoid and industrial robots.
Hebbian Robotics
United States
San Francisco YC S26 start-up founded 2026 that builds HFlow, an open-source SDK for checking and curating robot training data.
Kinetic Blocks
Norway
Oslo startup whose marketplace, in beta since Sept 2026, sells verified humanoid-robot training data, each dataset graded A+ to F by its KBQS system.
Tau Robotics
United States
Palo Alto AI robotics startup building general-purpose robot learning systems that enable robotic arms to acquire skills autonomously in the real world.
Hive Autonomy
Norway
Norwegian AI startup (est. 2020) retrofitting industrial machines like wheel loaders, forklifts, and excavators with autonomous AI technology.
NVIDIA Robotics
United States
NVIDIA robotics division delivering Isaac simulation and training tools, Jetson edge compute modules, and Omniverse digital twin frameworks for physical AI.
Attraclab
Japan
Attraclab is a Saitama, Japan field-robotics integrator building drones, UGVs and USVs with AI and GIS for agriculture, construction and logistics.
commissure Inc.
Japan
University of Tokyo spin-off founded 2023 in Tokyo making tactile sensing, haptic devices and touch-data tools for robot imitation learning; NEDO NEP pick 2026.
AceiiLab 一思智能
China
AceiiLab is a Shenzhen-founded startup making the AceiiLab A1 AI tennis-serving robot with dual-lens vision, founded by ex-DJI and KUKA engineers.
ROB'OCC
France
Brens, France maker founded 2023 of the ROC-E offline AMR (200 kg payload); one unit ran 4,127 missions in a year at MECAPREC.
Robotics needs a service framework.
RSF defines a common language for robot service capability, lifecycle operations, certification pathways, and service-provider networks.
Join the RobotToday community on LinkedIn
Daily robotics news, in-depth analysis, conference highlights, and discussions with professionals worldwide.
