Fourier Advances Humanoid Robotics with NVIDIA Isaac Gym for Real-World Applications

Shanghai-based robotics company Fourier is at the forefront of leveraging NVIDIA’s cutting-edge technology to enhance humanoid robots for real-world applications. According to the NVIDIA blog, the company recently expanded its GRx humanoid robot series with the introduction of the GR-2, which offers significant advancements in hardware design, adaptability, and agility.

Humanoid robot GR-2 developed with NVIDIA Isaac Gym

Fourier’s development of the GR-2 humanoid robot utilized NVIDIA Isaac Gym, a platform for reinforcement learning, to streamline the training process. This approach allows you to simulate complex real-world scenarios, reducing testing time and costs. The company is shifting its workflow to NVIDIA Isaac Lab, which provides an open source modular framework aimed at simplifying robot learning.

NVIDIA tools enable Fourier to simulate complex multi-robot scenarios and diverse environments to improve AI decision-making. The simulation includes pre-trained grasp algorithms, significantly reducing real-world trial and error, saving both time and resources.

AI optimization for real-world robotics

Fourier optimized GR-2’s AI capabilities using NVIDIA TensorRT for real-time inference optimization and the CUDA library for enhanced processing. This allows the robot to perform complex manipulations, such as transitioning from floor to stand, with a success rate of 89% after 3,000 repetitions, a significant improvement over existing methods.

The integration of these technologies will not only accelerate the training process, but also enhance the robot’s real-time motion control and AI-based decision-making, setting a new standard for human-robot interaction in a variety of industries, including healthcare and scientific research.

Explore next-generation robotic capabilities

Fourier adopted NVIDIA technology to reduce training times and improve simulation accuracy, fostering collaboration between the company’s engineering and R&D teams. This opens up new possibilities for complex AI capabilities, such as language models and predictive analytics, that were previously too resource-intensive to implement.

Fourier CEO Alex Gu highlighted these advancements, saying, “Improvements in real-time motion control and AI decision-making are setting new standards for humanoid robotics, especially in fields such as hospitality, academic research, and medical rehabilitation.”

For more information, see the NVIDIA blog.

Image source: Shutterstock