Skip to content

Google DeepMind Unveils Apollo: A Humanoid Robot That Folds Clothes and Sorts Items Using Natural Language Commands

Apollo marks a significant leap in robotics. While it's not yet at human-like levels, it's a step towards more intuitive robot-human interaction.

There is a poster in which there is a robot, there are animated persons who are operating the...
There is a poster in which there is a robot, there are animated persons who are operating the robot, there are artificial birds flying in the air, there are planets, there is ground, there are stars in the sky, there is watermark, there are numbers and texts.

Google DeepMind Unveils Apollo: A Humanoid Robot That Folds Clothes and Sorts Items Using Natural Language Commands

Google DeepMind has unveiled Apollo, a humanoid robot capable of performing tasks like folding clothes and sorting items using natural language commands. The robot's capabilities are the result of DeepMind's latest AI models, Gemini Robotics 1.5 and Gemini Robotics-ER 1.5.

Gemini Robotics-ER 1.5 specializes in understanding physical spaces and making logistical decisions. It works in tandem with Gemini Robotics 1.5, which turns visual information and instructions into motor commands for the robot.

The robot's capabilities are based on a defined set of rules and structured scenario planning, rather than independent thought. Researchers, including Zhong Lin Wang, are working on developing sensing technologies for robots to preserve a sense of touch and tactile feedback. However, robots are still far from having sensing and thinking capabilities on par with humans. They currently lack the ability to register pain and smell, which are important for uncertain environments.

A roadmap for AI in robotics was published in Nature Machine Intelligence by Aude Billard et al., outlining the progress and future directions in the field. The integration of large language models with robotics allows users to ask the robot to perform tasks using simple language.

While Apollo represents a significant advancement in robotics, it's important to note that robots are still not at the level of human-like sensing and thinking. Ongoing research, such as that led by Zhong Lin Wang, aims to bridge this gap by developing sensing technologies for robots. The future of AI in robotics, as outlined in the roadmap published by Aude Billard et al., promises exciting developments in the field.

Read also:

Latest