GR-3 is equipped with Fourier’s proprietary Full-Perception Multimodal Interaction System, integrating vision, audio, and tactile feedback into a unified real-time emotional processing engine.
Who said machines can’t feel? Fourier’s new “Care-bot” is here to listen, respond, and flip the script.
The robotics and AI company unveiled GR-3, its first full-size “Care-bot” and the latest addition to its GRx humanoid robot lineup.
Unveiled on August 6, GR-3 reimagines human-robot interaction through the lens of empathy, emotion, and connection.
“Love, above all functionality.” That’s the design philosophy guiding GR-3’s creation.
And it shows.
Care is the code
The robot is not only designed to perform tasks but to build emotional rapport in real-world environments, including public spaces, homes, hospitals, and eldercare facilities.
Whether offering comfort to a child, acting as a responsive companion to an older adult, or assisting with mobility and health monitoring, GR-3 brings a new dimension to what machines can do and feel.
Unlike traditional robots built purely for utility, the “Care-bot” emphasizes human-centered service where emotional interaction is just as important as mechanical precision.
The design backs up that ambition. With a soft-touch shell, warm neutral tones, and premium automotive upholstery, GR-3 has been built to feel familiar rather than foreign.
Fourier unveiled GR-3, its first full-size “Care-bot” and its latest GRx humanoid series entry. Credit – PRNewsfoto/Fourier Co. Ltd.
It stands 165 cm tall, weighs 71 kg, and boasts 55 degrees of freedom, allowing for balanced, natural movement.
Its animated facial interface and fluid gait work together to create a sense of personality and presence, rather than robotic detachment.
At the core of GR-3’s emotional intelligence is Fourier’s Full-Perception Multimodal Interaction System, which fuses sight, sound, and touch into a single, real-time emotional engine.
It can localize voices using a four-microphone array, make eye contact while you’re speaking, and recognize your face through structured-light cameras.
But it doesn’t stop there – 31 pressure sensors across its body detect touch, triggering lifelike responses like blinking, eye tracking, or subtle emotional gestures.
GR-3 doesn’t just hear and see you, it reacts like it actually gets you.
But GR-3 isn’t just built to detect. It’s built to think. What sets it apart is its dual-path response architecture.
“Fast thinking” handles reflexive, rule-based responses, while “slow thinking” draws on a large language model to carry out complex, contextual conversations.
This hybrid brain allows GR-3 to adapt not only to tasks but to tone, body language, and emotional cues.
GR-3 is designed to enable meaningful interactions in a wide range of real-world environments, including public services, academic research, clinical applications, and eventually, personal settings. Credit – PRNewsfoto/Fourier Co. Ltd.
“GR-3 interprets emotion, recognizes intent, and responds with awareness—bridging the gap between perception and connection.”
It also comes ready for the real world. The robot features a hot-swappable battery system, intelligent power management for continuous operation, and locomotion modes that allow it to squat, bend, and even adopt expressive gaits like “bouncy walk” or “fatigue mode” based on the environment.
Fourier isn’t positioning GR-3 as just a product; it’s laying the groundwork for a platform. Its modular structure, developer-friendly APIs, and support for third-party algorithms make it customizable for researchers, caregivers, and service industries alike.
GR-3 marks a new phase in Fourier’s pursuit of human-robot integration. Every aspect of its design—from hardware architecture to software framework—was shaped by a key question: Can a robot be genuinely accepted in human environments?
With GR-3, Fourier is betting the answer is yes—and that the future of robotics might just be a little more human.
MasterCard