Physicl has raised $10 million to build technology aimed at one of robotics’ biggest challenges: making the digital world accurate enough for robots to learn from it.
The company is working on the sim-to-real gap, the mismatch between what robots experience in simulation and what they encounter in the physical world. More realistic virtual environments can help robots practice tasks safely and repeatedly before being deployed in real-world settings.
Why this matters
Robotics development is often expensive, slow, and constrained by the difficulty of collecting real-world training data. If Physicl can turn digital product pipelines into realistic robot-training environments, it could make robotics experimentation faster and more accessible.
- Lower cost: Teams can test more scenarios virtually before building or deploying hardware.
- Greater safety: Robots can learn from simulated mistakes instead of dangerous real-world trials.
- Faster progress: More realistic simulations could speed up development across logistics, manufacturing, retail, and home robotics.
While the company is still early, this funding is a promising sign that investors see real potential in AI-powered simulation as a foundation for the next generation of useful robots.