Foucault Releases Universal Humanoid Robot GR-2, Kicking Off a New Journey for the GRx Series



Foucault's independently developed new generation of universal humanoid robot GR-2 has been launched with great fanfare. Based on the product vision of "creating the best embodied carrier for AI," the GR-2 has brought remarkable innovations and enhancements in multiple key aspects such as hardware, design, and development frameworks. It showcases more flexible, powerful, and open characteristics to meet the needs of developers in various fields and diverse application scenarios.

**Comprehensive design upgrade, a new milestone for the GRx series**
Image
The GR-2 continues the bionic design language of the GR-1 in appearance and further optimizes its body proportions. The upgraded model stands at 175cm tall and weighs 63kg, with a total of 53 degrees of freedom, a single arm motion load of 3kg, and the ability to perform more complex operations. The new generation of battery capacity has doubled, extending the endurance time to 2 hours. In addition to supporting direct charging, a detachable battery replacement scheme has been added to fully meet motion needs.

The entire machine layout of the GR-2 adopts an internal wiring design, placing cables inside the robot, which not only enhances the overall aesthetics but also simplifies modular adaptation, reducing obstacles for developers' peripheral device layout, and leaving ample space for scenario adaptation development. To optimize control strategies and performance, the joint layout of the GR-2 has been changed from parallel to series, allowing each joint's movement to be calculated independently, which is beneficial for developers to deploy control algorithms, facilitating debugging and maintenance, and reducing manufacturing costs.

**Human-like tactile dexterous hands for more flexible upper limb operations**

The flexibility and precision of upper limb operations on the GR-2 have been significantly improved, thanks to the iterative upgrade of the performance of the dexterous hands by Foucault. The new generation of dexterous hands equipped on the GR-2 increases the number of degrees of freedom from the initial 6 to 12, perfectly replicating the shape and size of the human hand, while simulating the natural movement of the human hand to the greatest extent.

**Foucault independently develops a new generation of dexterous hands**

The new generation of dexterous hands also features an array of 6 tactile sensors and motion control algorithms, capable of perceiving gripping force in real-time, adjusting grasping strategies, and achieving precise force control and grasping of objects of different shapes, materials, and weights.

As the next main battlefield for the humanoid robot industry to actively promote commercialization, upper limb operations still require a large amount of data training to achieve generalization and break through the barriers of specialized capabilities. The GR-2 supports various upper limb teaching methods such as VR remote control, drag teaching, and upper computer terminal control. It quickly creates actions, records multi-dimensional data such as upper limb motion trajectory, joint angles, force, dexterous hand movement, and tactile information, providing strong support for collecting more real training data and helping developers bridge the gap between simulation and reality.

**GR-2 achieves more flexible upper limb operations**

**Actuator 2.0 for stronger motion capabilities**

The GR-2 achieves a more anthropomorphic performance in appearance structure and motion control. To this end, Foucault has tailored 7 different models of FSA actuators for the robot's hip, ankle, and other joints to meet the torque requirements of each joint. These self-developed actuators not only perfectly adapt to the robot's bionic form but also effectively reduce hardware costs.

The new generation of actuators provides stronger power for the GR-2, with a maximum joint peak torque of over 380N.m. Combined with a comprehensive actuator control algorithm, the GR-2 has stronger load capacity and smooth motion control.

The new version of the actuator uses a dual encoder system, increasing the encoder accuracy by 2 times, providing more accurate absolute position and speed feedback, ensuring the control accuracy of the GR-2 during various actions and achieving high dynamic response.

**A more open development platform to accelerate the embodiment of AI**

Foucault places great emphasis on the needs and experiences of developers. In this product upgrade, the original development platform has been optimized to form a development interface scheme based on mainstream programming languages. Whether it is robot research scholars, industrial automation engineers, or AI software developers, they can quickly get started and efficiently customize a variety of applications, achieving robot technology innovation.

The scheme supports server-client model algorithm program development, encapsulates a series of concise and easy-to-use APIs, and integrates optimized algorithm modules such as machine vision, path planning, and force control feedback. It reduces the development threshold and simplifies the implementation process of complex tasks, significantly improving development efficiency. It currently supports open-source frameworks such as NVIDIA Isaac Lab, ROS, Mujoco, and Webots.

The platform provides detailed log recording and remote monitoring functions, built-in strict security protocols and fault recovery mechanisms to ensure safe and reliable development processes, and provides developers with comprehensive documentation and rapid technical support.

**Foucault GRx, born for AI**

For the future evolution of humanoid robot bodies, Foucault proposes the "product hexagon" concept, focusing on six dimensions including motion intelligence, dexterous operation, cognitive intelligence, bionic design, user experience, and commercial application, to clarify the capability standards and development direction of the robot body.

"We have been thinking during the product conception process, there are so many of the world's best minds exploring AI, but it still needs a physical container to carry it," said Gu Jie, the founder and CEO of Foucault. "When the humanoid robot body achieves optimality in all dimensions and becomes a hexagonal warrior, it will definitely be the best embodiment choice for AI. The AI here is not limited to large models but also includes the entire process of robot perception, decision-making, and execution."

The Foucault GRx series is committed to creating more anthropomorphic, versatile, and user-friendly bipedal humanoid robots for its cooperative ecosystem, including scientific research institutions, leading companies, scenario developers, and end customers. It promotes the close integration of AI with the physical world and achieves true symbiosis between humans and machines.

A total of 1 page 1 data