- Figure’s new humanoid robot accepts voice commands to assist with household chores.
- The device leverages advanced AI and voice recognition technology.
- This innovation represents a significant leap in home robotics, promising to simplify daily tasks.
Figure, an innovator in robotics technology, has unveiled its latest humanoid robot, which can now accept voice commands to carry out a range of household tasks.
Figure founder and CEO Brett Adcock announced the Bay Area robotics firm’s decision to step away from an OpenAI collaboration, which is centered around Helix, a “generalist” Vision-Language-Action (VLA) model.
VLAs are a new phenomenon for robotics, leveraging vision and language commands to process information.
The new device uses advanced AI-driven voice recognition to help with chores, promising to transform the way users manage daily activities at home.
“Helix displays strong object generalization, being able to pick up thousands of novel household items with varying shapes, sizes, colors, and material properties never encountered before in training, simply by asking in natural language.” Figure quotes.
Upon receiving a voice command, the robot visually assesses the environment and performs the task, thus bridging the gap between vision and language processing.
During a live demonstration, the robot accurately followed various voice commands to perform tasks such as cleaning, organizing and even assisting with meal preparation.
Figure is showcasing the Vision-Language Model (VLM) by highlighting its work with the O2 humanoid robot in residential settings. Households also present a significant challenge for robots due to their lack of consistent structure found in warehouses and factories.
The robot incorporates machine learning algorithms that enable it to adapt to each household’s unique needs over time.
“For robots to be useful in households, they will need to be capable of generating intelligent new behaviors on-demand, especially for objects they’ve never seen before,” Figure says.
“Teaching robots even a single new behavior currently requires substantial human effort: either hours of PhD-level expert manual programming or thousands of demonstrations.”
Manual programming isn’t feasible for home robotics, as the vast variability in domestic environments—from differing layouts and tools to unpredictable changes in furniture and lighting—renders such an approach too time-consuming and costly.
Instead, extensive training involving hundreds of hours of repetition is required to equip robots with the reliability needed to perform highly variable tasks in these dynamic settings.
Edited By Annette George
