Spun out of Carnegie Mellon’s Safe AI Lab, the start-up puts robots through their paces in thousands of scenarios involving virtual humans who stumble, crouch or carry loads.
Safeworld came out of the shadows on October 5 with more than $12 million in seed funding, led by Shine Capital and a16z Speedrun, alongside Box Group, Carnegie Mellon University Endeavors, and SV Angel. It was founded by Ding Zhao, director of Carnegie Mellon's Safe AI Lab, Kyle Wong, experienced start-up manager, and Simo Rachidi, a machine learning engineer.
Digital Humans Facing Robots
The company is tackling a problem that is growing with the advent of the general AI in the control of robots: how to ensure that they will not hurt anyone. Its platform evaluates control systems in simulators like Genesis or MuJoCo, placing human models in them. It takes on thousands of situations, including borderline cases where a person stumbles, squats or transports an object.
Safeworld targets manufacturers who deploy autonomous robots in unstructured environments, such as factories or construction sites. Gritt Robotics, which designs the "brain" of solar panel installation robots, is among its first partners. Investors believe that safety standards must be established before the proliferation of deployments leads to accidents. The economic model still hesitates between software platform and service delivery.
Read the full article on the original source
Go to the source →