Robot studies how to be human from language, emotions
Imagine what it would be like to have a robot friend that can do things like take selfies, toss a ball, eat popcorn and play air guitar?
Well, you might not have to wait too long.
Researchers at the University of Tokyo have created a robot that can do all that and more, thanks to the power of GPT-4, the latest and most advanced large language model (LLM) in the world.
A researcher gives Alter3, a humanoid robot, verbal instructions. (University of Tokyo)
What is the Alter3 humanoid robot, how does it work?
Alter3 is a humanoid robot that was first introduced in 2016 as a platform for exploring the concept of life in artificial systems. It has a realistic appearance and can move its upper body, head and facial muscles with 43 axes controlled by air actuators. It also has a camera in each eye that allows it to see and interact with humans and the environment.
WHAT IS ARTIFICIAL INTELLIGENCE (AI)?
Alter3 interacts with a human. (University of Tokyo)
But what makes Alter3 really special is that it can now use GPT-4, a deep learning model that can generate natural language texts from any given prompt, to control its movements and behaviors. This means that instead of having to program every single action for the robot, the researchers can simply give it verbal instructions and let GPT-4 generate the corresponding Python code that runs the Android engine.
CALIFORNIA LEGISLATIVE SESSION TO BE DOMINATED BY AI REGULATIONS AND STATE'S STRUGGLING BUDGET
For example, to make Alter3 take a selfie, the researchers can say something like:
"Create a big, joyful smile and widen your eyes to show excitement. Swiftly turn the upper body slightly to the left, adopting a dynamic posture. Raise the right hand high, simulating a phone. Flex the right elbow, bringing the phone closer to the face. Tilt the head slightly to the right, giving a playful vibe."
And GPT-4 will produce the code that makes Alter3 do exactly that.
Alter3 mimics taking a selfie. (University of Tokyo)
MORE: HUMANOID ROBOTS ARE NOW DOING THE WORK OF HUMANS IN A SPANX WAREHOUSE
What can the Alter3 humanoid robot do with GPT-4?
The researchers have tested Alter3 with GPT-4 in various scenarios, such as tossing a ball, eating popcorn, and playing air guitar. They have also experimented with different types of feedback, such as linguistic, visual, and emotional, to improve the robot’s performance and adaptability.
Alter3 mimics playing a guitar. (University of Tokyo)
One of the most interesting aspects of Alter3’s behavior is that it can learn from its own memory and from human responses. For instance, if the robot does something that makes a human laugh or smile, it will remember that and try to repeat it in the future. This is similar to how newborn babies imitate their parents’ expressions and gestures.
Alter3 mimics jogging. (University of Tokyo)
MORE: THE NEXT GENERATION OF TESLA'S HUMANOID ROBOT MAKES ITS DEBUT
The researchers have also added some humor and personality to Alter3’s actions. In one case, the robot pretends to eat a bag of popcorn, only to realize that it belongs to the person sitting next to it. It then shows a surprised and embarrassed expression and apologizes with its arms.
Alter3, the humanoid robot (University of Tokyo)
Why is this humanoid robot AI important and what are the implications?
The research team behind Alter3 believes that this is a breakthrough in the field of robotics and artificial intelligence, as it shows how large language models can be used to bridge the gap between natural language and robot control. This opens up new possibilities for human-robot collaboration and communication, as well as for creating more intelligent, adaptable, and personable robotic entities.
Alter3 mimics seeing a pretend snake. (University of Tokyo)
MORE: HOW THIS ROBOT HELPS YOU PROTECT AND CONNECT YOUR HOME
The paper, titled "From Text to Motion: Grounding GPT-4 in a Humanoid Robot ‘Alter3,’" was written by Takahide Yoshida, Atsushi Masumori and Takashi Ikegami and is available on the preprint server arXiv. The authors hope that their work will inspire more research and development in this direction and that one day we might be able to have robot friends that can understand us and share our interests and emotions.
Kurt's key takeaways
Alter3 is an example of how natural language processing and robotics can work together to create pretty incredible interactions. By using GPT-4, the robot can perform a variety of tasks and behaviors based on verbal commands, without requiring extensive programming or manual control. This also allows the robot to learn from its own experience and from human feedback and to express some humor and personality. Alter3 demonstrates the potential of large language models to improve the field of robotics and artificial intelligence as well as bring us closer to having robot friends that can relate to us and entertain us.
What do you think of Alter3 and its abilities? Would you like to have a robot like that in your life? Let us know by writing us at Cyberguy.com/Contact.
For more of my tech tips & security alerts, subscribe to my free CyberGuy Report Newsletter by heading to Cyberguy.com/Newsletter.
Ask Kurt a question or let us know what stories you'd like us to cover.
Answers to the most asked CyberGuy questions:
- What is the best way to protect your Mac, Windows, iPhone and Android devices from getting hacked?
- What is the best way to stay private, secure and anonymous while browsing the web?
- How can I get rid of robocalls with apps and data-removal services?
Ideas for using those Holiday Gift cards:
Copyright 2024 CyberGuy.com. All rights reserved.
Kurt "CyberGuy" Knutsson is an award-winning tech journalist who has a deep love of technology, gear and gadgets that make life better with his contributions for Fox News & FOX Business beginning mornings on "FOX & Friends." Got a tech question? Get Kurt’s CyberGuy Newsletter, share your voice, a story idea or comment at CyberGuy.com.