Figure’s humanoid robot takes voice orders to help around the house

Figure reveals Helix, a VLA model for humanoid robots to perform home tasks.

: Brett Adcock of Figure announced Helix, a Vision-Language-Action model for humanoid robots. Helix combines visual data and language prompts for real-time task execution in homes. It's designed to handle various household objects with natural language commands. Figure aims to prioritize home robotics despite the current focus on workplace applications.

Brett Adcock, founder and CEO of Figure, has unveiled Helix, a new Vision-Language-Action (VLA) model for humanoid robots designed to assist with household tasks. Helix operates by combining visual data with natural language prompts to control the robot in real time, allowing it to handle a wide array of novel household items based on simple commands.

Figure showcases Helix's capabilities in tandem with its 02 humanoid robot, aiming to overcome the challenges posed by the less structured environment of homes compared to warehouses or factories. Despite the high costs and difficulties associated with home robotics, Figure plans to make the domestic setting a priority for their humanoid robot applications.

The current focus on workplace pilots, exemplified by collaborations with companies like BMW, will gradually shift towards domestic uses. Training these robots for home environments remains at an early stage, with the announcement serving as a recruitment effort to attract more engineers to help develop and expand the project.

Helixhumanoid roboticsAdcock