Generative AI models can produce images in response to prompts within seconds, and they’ve recently been used for everything from highlighting their own inherent bias to preserving precious memories.
Now, researchers from Stephen James’s Robot Learning Lab in London are using image-generating AI models for a new purpose: creating training data for robots. They’ve developed a new system, called Genima, that fine-tunes the image-generating AI model Stable Diffusion to draw robots’ movements, helping guide them both in simulations and in the real world. The research is due to be presented at the Conference on Robot Learning (CoRL) next month.
The system could make it easier to train different types of robots to complete tasks—machines ranging from mechanical arms to humanoid robots and driverless cars. It could also help make AI web agents, a next generation of AI tools that can carry out complex tasks with little supervision, better at scrolling and clicking, says …