Home / Technology / Robot Training: Virtual Worlds Unlock Real-World Skills
Robot Training: Virtual Worlds Unlock Real-World Skills
15 Oct
Summary
- DexForce uses virtual simulations for robot training.
- Synthetic data aims to cut costs and boost efficiency.
- Company secured 1 billion yuan in recent funding.
DexForce, a Shenzhen-based startup, is pioneering a new approach to humanoid robot training through advanced virtual simulations. This innovative method utilizes synthetic data generated in digital environments, aiming to overcome the significant challenge of teaching robots diverse real-world tasks without constant human supervision.
The company's strategy focuses on creating digital replicas of environments and tasks, allowing robots to learn and adapt within these simulations before transferring knowledge to physical machines. This virtual-to-real-world process is reportedly faster and more cost-effective than traditional methods reliant on extensive real-world data collection.
Founded in 2021, DexForce has attracted substantial investment, recently raising 1 billion yuan ($147 million) in a Series B funding round. The company's software is utilized by major global manufacturers, and its humanoid robots are already performing tasks like making coffee in various cities across China.
