SUMMARY:

 

Roboto 01 (also published as Roboto Origin) is an open-source, DIY bipedal humanoid robot developed by Shanghai RoboParty Technology Co., Ltd.

 

Standing 125 cm tall and weighing 34 kg, it serves as a standardized reference platform for universities, physical AI researchers, and developers.

 

It features 23 degrees of freedom, an onboard D-Robotics RDK X5 compute board (10 TOPS), and custom USB-to-CAN motor control pipelines.

 

EDITORIAL:

 

RoboParty designed the Roboto 01 as an accessible, fully transparent hardware-software ecosystem.

 

Rather than relying on closed intellectual property, the platform makes its full CNC CAD blueprints, PCB schematics, IsaacLab reinforcement learning pipelines, and ROS 2 control nodes available in open repositories.

 

It bridges the gap between simulation and real-world hardware, offering quick Sim2Sim and Sim2Real workflows for dynamic bipedal locomotion.

 

SPECIFICATIONS:

 

Dimensions & Build

Height: 125 cm (1.25 m / 3.9 ft)

Weight: 34 kg (75 lbs)

Chassis: Full CNC-machined aluminum frame with internal cable routing and lightweight housing panels

Use Cases: Research laboratories, university education, embodied AI benchmarking, and open-source locomotion algorithm development

Mobility & Performance

Degrees of Freedom (DOF): 23 DOF total

Legs: 12 DOF (6 DOF per leg for spatial movement)

Arms: 10 DOF (5 DOF per arm)

Torso/Waist: 1 DOF

Top Speed: Up to 3.0 m/s (10.8 km/h)

Locomotion Capability: Dynamic walking, running, stair navigation, jumping, and push-recovery balance algorithms

Actuation & Manipulation

Actuators: DAMIAO servo motors paired with high-torque planetary reducers

Peak Torque:

Leg Joints: Up to 120 Nm peak torque

Arm Joints: Up to 27 Nm peak torque

End-Effectors: Forearm mount points compatible with custom or developer-added dexterous effectors

Perception & Compute

Vision: Support for optional Intel RealSense D435i RGB-D depth camera or E1R LiDAR

Telemetry: Integrated high-precision IMU, 6-axis gyroscope, and real-time encoder feedback over CAN FD

AI Compute Board: D-Robotics RDK X5 (8× ARM Cortex-A55 @ 1.5 GHz CPU, 10 TOPS BPU AI unit)

Software Stack: Ubuntu Linux, ROS 2 middleware, MuJoCo simulation environment, and IsaacLab RL training pipeline

Power & Interfaces

Battery Pack: 48V 15Ah lithium battery pack

Connectivity: USB-to-CAN, CAN FD interface, Gigabit Ethernet, Wi-Fi 6, and Bluetooth 5.4

 

Contact for current price.

Image: Humanoid Press

 

Company Source:

https://www.youtube.com/shorts/JlrvWcpRiIE