Disassembling the Yu Shu G1 Robot: How is the control system designed?

Source: Zhinew Automotive

The China Post Securities Electric New Team (Su Qianye / Sheng Wei / Yang Shuaibo) has produced an in-depth analysis report, noting that Yushu Technology shipped over 5,500 units throughout the year, ranking first globally in shipment volume — this mass-produced model.

This mainly concerns the controller section.

◎ Regarding the physical layout and protection of the controller, the control board is placed in the robot’s back area, partially covered with heat dissipation fins and supplemented with air cooling design to ensure thermal stability of core components under high load.

To provide comprehensive physical protection, a triple protection system is built behind the control board: the outermost layer is a sturdy robot back shell, the middle layer is filled with cushioning foam to absorb shocks, and the innermost layer features a plastic protective layer. This multi-layered protection structure effectively safeguards the core control unit in complex operational environments.

◎ The main control board’s core computing unit uses the Rockchip RK3588S chip, manufactured with 8nm process technology, featuring an 8-core CPU (including 4 Cortex-A76 large cores and 4 Cortex-A55 small cores), integrated with a Mali G610MC4 GPU and a NPU with 6 TOPS computing power, providing strong support for the robot’s basic operation and control.

◎ In terms of storage configuration, the mainboard is equipped with two BIWIN BMW2CX32H2A-32G-X memory chips, totaling 8GB of RAM, and one FORESEE FEMDNN064G-58AAA storage chip, offering 64GB of storage space, ensuring smooth system operation and efficient data read/write.

◎ Regarding chip information protection, disassembly reveals that, besides the main control, memory, and storage chips, the model numbers and information of other key chips on the board are deliberately erased or hidden.

◎ In version differentiation and interaction control, the EDU version (high-end model) differs from the low-end version mainly by the addition of an Nvidia Jetson Orin NX chip. This high-performance module can provide up to 100 TOPS of edge AI computing power, enabling stronger deep learning and environmental perception capabilities.

◎ The accompanying remote control supports Bluetooth connection, with a transmission distance exceeding 100 meters and a battery life of up to 4.5 hours; its unique stretchable design allows users to mount a phone as a real-time monitoring screen. By installing a dedicated app, users can achieve visual remote control and status monitoring of the robot.

Summary: The above is for electronic component reference.

Special statement: The above content only reflects the author’s personal views or positions and does not represent Sina Finance Headlines’ opinions or stances. If you need to contact Sina Finance Headlines regarding copyright or other issues related to this work, please do so within 30 days of the above content’s publication.