Humanoid-universe.com

Appreciate the detailed answer. Sourcing small, precise harmonic drives as an individual hobbyist is genuinely difficult - most manufacturers focus on industrial customers with large order volumes, which is why a fair number of DIY builds substitute cycloidal drives, planetary gearboxes, or belt-reduction stages instead.

Thanks for laying this out, genuinely useful. Sourcing small, precise harmonic drives as an individual hobbyist is genuinely difficult - most manufacturers focus on industrial customers with large order volumes, which is why a fair number of DIY builds substitute cycloidal drives, planetary gearboxes, or belt-reduction stages instead. Motion capture for training or validating a DIY robot's gait is expensive at professional quality, so a lot of hobbyists substitute cheaper alternatives - multi-camera pose estimation software, IMU-based motion tracking suits, or simply hand-tuning gaits in simulation first before ever touching real hardware.

@gary.tanaka2 This matches what I've seen too. The Berkeley Humanoid Lite stands about 1 meter tall and weighs roughly 16 kg, which is a genuinely reasonable target size for a first serious hobbyist build - big enough to be a real bipedal platform, small enough to not need industrial-scale actuators or a huge budget.

From hands-on experience, A realistic hobbyist bill-of-materials for a small biped balloons fast once you add a real IMU, decent motor controllers, a reasonable battery/BMS setup, and enough compute for onboard control - the headline '$5k' budget assumes a fair amount of DIY fabrication rather than buying finished modules. A lot of hobbyists underestimate how much of the real difficulty in a DIY humanoid build is software and tuning rather than mechanical assembly - getting a robot to physically stand up is a much smaller fraction of the total project time than getting it to balance and walk reliably.

@harmonicjen60 This lines up with my experience. The Berkeley Humanoid Lite stands about 1 meter tall and weighs roughly 16 kg, which is a genuinely reasonable target size for a first serious hobbyist build - big enough to be a real bipedal platform, small enough to not need industrial-scale actuators or a huge budget.

@diego.moore6 Minor factual note: Building a scoped-down project first - a single leg, an arm, or a torso-only upper body - is a common and sensible way to learn the mechanical and control fundamentals before committing to the cost and complexity of a full bipedal platform. Group-buy arrangements for harder-to-source parts (small harmonic drives, precision bearings, custom PCBs) are a common way hobbyist communities work around the fact that specialty components are often only sold in industrial minimum-order quantities.

Here's the relevant bit as far as I understand it: A first successful unassisted walk, from a standing start on a from-scratch project, commonly takes many months of part-time weekend work even for someone with a solid mechanical and software background - it's a genuinely hard milestone, not a weekend project.

@choi98 That's the official framing, at least - reality tends to lag a bit. Starting in simulation (even a simplified physics sim of just the leg or the target gait) before touching real hardware is one of the more reliable ways to avoid destroying expensive parts while debugging balance and walking control from scratch. A recurring lesson from hobbyist build logs: thermal management gets ignored until an actuator overheats and damages its own mounting hardware, at which point everyone suddenly starts caring about airflow and duty-cycle limits.

Counterpoint: A lot of hobbyists underestimate how much of the real difficulty in a DIY humanoid build is software and tuning rather than mechanical assembly - getting a robot to physically stand up is a much smaller fraction of the total project time than getting it to balance and walk reliably.

@choi98 This matches what I've seen too. Jetson-class boards remain a common choice for onboard compute on hobbyist builds needing real vision or learned-policy inference, while a Raspberry Pi is often adequate for lower-level joint control and telemetry if the heavier compute is offloaded to a base station.