Two open-source humanoid robots
Asimov 1 by Menlo Research, and a second humanoid built on RoboParty's open Roboto_origin design, are real humanoid robots whose designs are published. Their links and joints are modelled from each robot's own URDF. You learn their actuators, route their buses and daisy chains, wire their harnesses by hand, fault-find a robot that will not stand, and walk them in physics in your browser. Both open in Robot Studio. Their courses are never sold on their own: they come free with Premium.
Guided builds
Asimov 1, part by part
Six builds from the feet to the head, in the order of Menlo Research's assembly manual. Watch each step rendered in HD, then switch to 3D and turn the parts yourself.
[01] Step by step
How it goes
- [01]Meet the actuators, up close in 3D
- [02]Route the buses and the cable list
- [03]Wire the harness and pass the wiring assessment
- [04]Fault-find what stops it standing
- [05]Walk it in physics with its makers' trained policies
[02] Specification
What the courses cover
| Asimov 1 | Introduction to Humanoid Robotics, and Advanced Humanoid Engineering |
|---|---|
| Open-Source Humanoid | Build, Wire and Walk a Robot, on RoboParty's open Roboto_origin design |
| Model | Links and joints transcribed from each robot's URDF, in metres |
| Walking | Trained policies in real physics in the browser, and reinforcement learning on Asimov 1 |
| Assessment | Wiring assessments graded on the harness you build |
| Price | Free with Premium: built on open-source robots, never sold on their own |
| Notice | Roboto_origin is RoboParty's design; this course is not affiliated with or endorsed by RoboParty |
[03] Courses
Courses
Robotics
[01]Open-Source Humanoid: Build, Wire and Walk a Robot
A complete, hands-on look inside a real open-source humanoid. Every part of this robot is published: its mechanical design, its circuit board, its wiring, and the trained AI policies that make it walk. This course takes you through all of it, built on RoboParty's open Roboto_origin design. You take the robot apart in 3D, joint by joint and part by part: its 23 joints, the two kinds of Damiao motors that drive them, the main board that powers them and talks to them over four CAN buses, the battery, the E-stop and the computer. You learn how it is assembled, step by step, and you wire its harness by hand in the STEM Engine, graded against the published wiring guide. Then you make it move. You play the motions its makers recorded, and you run their own trained policies in real physics in your browser: walk it, steer it, push it, and watch it dance and get up off the floor. **Free with Premium:** this course is never sold on its own. It is built on an open-source robot, so it comes free with Premium: every Premium member can take it. Roboto_origin is RoboParty's design, published under CERN-OHL-W-2.0 (hardware) and GPL-3.0 (software). This course is CognitaSTEM's own and is not affiliated with or endorsed by RoboParty. The real robot is not for under-18s: this course is taught entirely in simulation.
42 lessons, 1 free to try · Free with Premium
Open a free lesson →Robotics
[02]Asimov 1: Advanced Humanoid Engineering
Take an open-source humanoid from CAD to a walking policy. Asimov 1 by Menlo Research is fully published: the mechanical design, the PCBs, the harness, the simulation model and the reinforcement-learning recipe that taught it to walk. This course goes through all of it, deeply. You'll learn how its 165 parts are machined, bent, printed or bought; how its actuators are sized; how its four circuit boards work, reverse-engineered from their own design files; how the 491 Wh battery is protected; and how to source, build and commission the robot. Then the software and the learning: the control loop, Menlo's PPO + AMP training recipe term by term, and the **Training Bay**, where Menlo's real trained policy walks Asimov in full physics in your browser, and where you can run and grade your own. Wire and repair the harness in the STEM Engine, train and upload a policy, and walk the 3D school as your own customised Asimov. **Free with Premium:** this course is never sold on its own. It is built on an open-source robot, so it comes free with Premium, with the Training Bay and an Asimov Hub avatar. Asimov 1 by Menlo Research is licensed CERN-OHL-S-2.0; Menlo's policy and training code are BSD-3-Clause.
71 lessons, 1 free to try · Free with Premium
Open a free lesson →Robotics
[03]Asimov 1: Introduction to Humanoid Robotics
Asimov 1 is Menlo Research's open-source humanoid: 1.25 m tall, about 32 kg, 25 actuators on CAN buses, a Radxa CM5 motion computer and a Raspberry Pi 5 in the head. Everything about it is published, from the CAD to the cable list. In this course you take the robot apart in 3D, module by module: the links and joints, the five actuator families, the compute and power stack, the CAN buses and the harness. Every module has an interactive Asimov Lab where you pose joints, play motions and inspect the real parts, and the lab checks your work as you go. Every part in the robot then gets its own lesson in 2D and 3D: its connectors, and how it is configured and used, from flashing the two computers to assigning actuator IDs, calibrating and the first start, following Menlo Research's own setup guides and videos. You finish by wiring the robot's harness by hand in the STEM Engine, graded against the published cable list, and configuring your own Asimov in 3D. This is the introduction. To go further, from PCB design and building your own Asimov to training its walking policy with reinforcement learning in the Training Bay, continue with **Asimov 1: Advanced Humanoid Engineering**. **Free with Premium:** this course is never sold on its own. It is built on an open-source robot, so it comes free with Premium: every Premium member can take it. Asimov 1 by Menlo Research is licensed CERN-OHL-S-2.0.
51 lessons, 1 free to try · Free with Premium
Open a free lesson →
[04] More builds
Build something else
Start today
Cut your first plate today
Create an account and open the Workshop in your browser, then take a short course to go further.