HomeBrew Back home
About Parts Part I — Earth Part II — Water Part III — Fire Part IV — Air Part V — Void
Syllabus → GitHub →
About Parts Earth Water Fire Air Void
Start Here Syllabus → GitHub →

Metal2Humanoid

A NAND2Tetris-style path through robotics: from materials → actuators → motion → behavior → learning → human meaning.

Five parts. Five elements. Each part contains two lectures, each with expandable units.
In collaboration with Menlo Research.
Browse the Parts ↓

About

This course is structured as five parts inspired by “elements” and a full-stack robotics build path. Each part contains two lectures with expandable units and project overviews.

(Work in progress) Unit content is placeholder for now: ⚙️ In Progress.

Parts

Start at Earth and climb toward Void: from grounded engineering reality to human questions of meaning.

Part I — Earth: Metal → Actuation

≈4 hrs Lectures 1–2 Projects 1–2

Motors, torque, heat, gearing, sensing, drivers, and digital twins — the unit cell of embodiment.

Open Part I →

Part II — Water: Actuation → Motion

≈4 hrs Lectures 3–4 Projects 3–4

Humanoid morphology, URDF/CAD, IK, trajectories, stability, and compliant control.

Open Part II →

Part III — Fire: Motion → Behavior

≈4 hrs Lectures 5–6 Projects 5–6

Skills, manipulation, behavior composition, planning, and robust recovery.

Open Part III →

Part IV — Air: Behavior → Learning

≈4 hrs Lectures 7–8 Projects 7–8

Imitation + RL, sim2real, diffusion policies, and VLA instruction-following.

Open Part IV →

Part V — Void: Learning → Humanoid

≈4 hrs Lectures 9–10 Projects 9–10

Meaning, ethics, labor, safety, openness, and final integration into a cohesive humanoid system.

Open Part V →

Part I — Earth: Metal → Actuation

Lectures: 1–2 · Projects: 1–2 · Total: ≈4 hrs

“The engineer’s first problem in any design situation is to discover what the problem really is.”

— Samuel Florman

Lectures & Units (Expandable)

Lecture 1 — Build a Motor ≈2 hrs · Project 1 Expand ▾
  • 1.1 What a motor really is (electromagnetism intuition) ⚙️ In Progress
    Placeholder: canonical intro + mental model + what to measure.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Paper, chapter, GitHub, doc, etc.
  • 1.2 Torque, speed, power (engineering intuition) ⚙️ In Progress
    Placeholder: back-of-the-envelope sizing + power limits.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Specsheets, calculators, or book chapters.
  • 1.3 Motor types (DC, BLDC, stepper) ⚙️ In Progress
    Placeholder: tradeoffs for robotics (control, torque density, cost).
    Open Video ↗ Paste a URL later.
    Open Link ↗ Comparisons, application notes, guides.
  • 1.4 Heat, efficiency, and failure modes ⚙️ In Progress
    Placeholder: thermal intuition, derating, smoke-test avoidance.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Thermal notes, motor constants, efficiency curves.
  • 1.5 Project 1 Overview: build + characterize a motor ⚙️ In Progress
    Placeholder: BOM, tools, steps, what charts/results to produce.
    Open Video ↗ Build walkthrough video.
    Open Link ↗ Repo folder for Project 1.
  • 1.6 Perspective: why actuators are only one bottleneck ⚙️ In Progress
    Placeholder: systems view + what's coming in later parts.
    Open Video ↗ Short talk / framing clip.
    Open Link ↗ Essay or design note.
Lecture 2 — What is an Actuator ≈2 hrs · Project 2 Expand ▾
  • 2.1 Transmissions: gearboxes & reduction ⚙️ In Progress
    Placeholder: harmonic vs planetary vs cycloidal, backlash, efficiency.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Application notes + teardown references.
  • 2.2 Bearings, shafts, fasteners, tolerances ⚙️ In Progress
    Placeholder: fits, alignment, wear, why things wobble.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Mechanical design references.
  • 2.3 Sensing: encoders, current sensing, torque estimation ⚙️ In Progress
    Placeholder: what sensors you need and what you can estimate.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Encoder docs + motor control notes.
  • 2.4 Drivers, ESCs, wiring, and safety ⚙️ In Progress
    Placeholder: power budgets, connectors, noise, grounding, fuses.
    Open Video ↗ Paste a URL later.
    Open Link ↗ ESC docs + wiring guides.
  • 2.5 CAD + digital twin basics (actuator module) ⚙️ In Progress
    Placeholder: model your actuator, export URDF, sanity-check ranges.
    Open Video ↗ Paste a URL later.
    Open Link ↗ CAD files + URDF template repo.
  • 2.6 Project 2 Overview: actuator module + test rig ⚙️ In Progress
    Placeholder: build plan + test plan + logging + deliverables.
    Open Video ↗ Build walkthrough video.
    Open Link ↗ Repo folder for Project 2.

Part II — Water: Actuation → Motion

Lectures: 3–4 · Projects: 3–4 · Total: ≈4 hrs

“Be formless, shapeless—like water.”

— Bruce Lee

Lectures & Units (Expandable)

Lecture 3 — Humanoid Morphology ≈2 hrs · Project 3 Expand ▾
  • 3.1 Kinematic chains: arms, legs, spine, and head ⚙️ In Progress
    Placeholder: chain structure, redundancy, reachability.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Kinematics notes / textbook chapter.
  • 3.2 DOF, link lengths, and why bodies look the way they do ⚙️ In Progress
    Placeholder: human vs robot morphology, tradeoffs, and constraints.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Reference designs / CAD / papers.
  • 3.3 CAD → URDF: how to build a usable robot model ⚙️ In Progress
    Placeholder: frames, joint limits, inertias, collisions, visuals.
    Open Video ↗ Paste a URL later.
    Open Link ↗ URDF docs, examples, templates.
  • 3.4 Center of Mass: balance and stability (intuitions that matter) ⚙️ In Progress
    Placeholder: CoM, support polygon, ZMP intuition, tipping.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Balance basics notes.
  • 3.5 Project 3 Overview: assemble the body model + sanity checks ⚙️ In Progress
    Placeholder: build your humanoid skeleton in CAD/URDF, confirm ranges and collisions.
    Open Video ↗ Project walkthrough.
    Open Link ↗ Repo folder for Project 3.
Lecture 4 — Motion & Control ≈2 hrs · Project 4 Expand ▾
  • 4.1 Forward and Inverse Kinematics: numerical IK, constraints, and singularities ⚙️ In Progress
    Placeholder: damped least squares, joint limits, redundancy.
    Open Video ↗ Paste a URL later.
    Open Link ↗ IK library docs / examples.
  • 4.2 Trajectories: interpolation, timing, limits, jerk ⚙️ In Progress
    Placeholder: smooth motion, constraints, safe parameterizations.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Trajectory generation references.
  • 4.3 Control: PID vs impedance, position vs torque ⚙️ In Progress
    Placeholder: stability intuition, compliance, safe interaction.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Control references (practical).
  • 4.4 Stability and compliance: stepping, stance, and constraint handling ⚙️ In Progress
    Placeholder: stable sequences, constraints, contact transitions.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Walking basics / balance control notes.
  • 4.5 Project 4 Overview: make the body move smoothly (IK + trajectories) ⚙️ In Progress
    Placeholder: implement IK + trajectory controller in sim (or hardware limb), show smooth motion.
    Open Video ↗ Project walkthrough.
    Open Link ↗ Repo folder for Project 4.

Part III — Fire: Motion → Behavior

Lectures: 5–6 · Projects: 5–6 · Total: ≈4 hrs

“Behavior is always directed toward something.”

— Niko Tinbergen

Lectures & Units (Expandable)

Lecture 5 — Manipulation Skills ≈2 hrs · Project 5 Expand ▾
  • 5.1 What is a “skill”? (interfaces, parameters, success conditions) ⚙️ In Progress
    Placeholder: skills as reusable behaviors with contracts and metrics.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Skill library docs, examples, or notes.
  • 5.2 Grasping & contact basics ⚙️ In Progress
    Placeholder: grasping fundamentals, contact forces, friction cones.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Hardware refs, gripper design notes, etc.
  • 5.3 State machines / behavior trees ⚙️ In Progress
    Placeholder: composing behaviors, FSM vs BT patterns, reactive control.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Skill implementations or tutorials.
  • 5.4 Robustness and Recovery ⚙️ In Progress
    Placeholder: what skills need to do to be robust and recover from failures.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Robustness and recovery notes.
  • 5.5 Project 5 Overview ⚙️ In Progress
    Placeholder: implement manipulation skills with state machines and recovery.
    Open Video ↗ Paste a URL later.
    Open Link ↗ FSM/BT patterns, recovery templates.
Lecture 6 — Task and Motion Planning ≈2 hrs · Project 6 Expand ▾
  • 6.1 Classical planning intuition ⚙️ In Progress
    Placeholder: define tasks as objectives + constraints + measurable outcomes.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Task spec templates, examples.
  • 6.2 Task & motion planning ⚙️ In Progress
    Placeholder: combining symbolic reasoning with continuous motion planning.
    Open Video ↗ Paste a URL later.
    Open Link ↗ BT/FSM examples and templates.
  • 6.3 Symbolic vs continuous ⚙️ In Progress
    Placeholder: discrete decisions vs continuous execution, bridging the gap.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Planning notes / classic references.
  • 6.4 Failure recovery ⚙️ In Progress
    Placeholder: detecting failures, replanning, recovery strategies.
    Open Video ↗ Paste a URL later.
    Open Link ↗ TAMP references + example repos.
  • 6.5 Project 6 Overview: a multi-step behavior with recovery ⚙️ In Progress
    Placeholder: compose 3–6 skills into one task (e.g., pick → carry → place) with recovery.
    Open Video ↗ Project walkthrough.
    Open Link ↗ Repo folder for Project 6.

Part IV — Air: Behavior → Learning

Lectures: 7–8 · Projects: 7–8 · Total: ≈4 hrs

“We can only see a short distance ahead, but we can see plenty there that needs to be done.”

— Alan Turing

Lectures & Units (Expandable)

Lecture 7 — Learning to Walk ≈2 hrs · Project 7 Expand ▾
  • 7.1 Why learning? limits of hand coding ⚙️ In Progress
    Placeholder: when to use learning vs hand-coded behaviors.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Dataset schemas, logging templates.
  • 7.2 Imitation learning ⚙️ In Progress
    Placeholder: behavior cloning, demonstration collection, covariate shift.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Teleop stack examples.
  • 7.3 Reinforcement learning basics ⚙️ In Progress
    Placeholder: rewards, value functions, policy gradient methods.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Benchmark/eval guides.
  • 7.4 Simulation → reality gap ⚙️ In Progress
    Placeholder: sim2real transfer, domain randomization, reality gap challenges.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Vision encoder references.
  • 7.5 Project 7 Overview ⚙️ In Progress
    Placeholder: what “supervision” means in robotics and where it comes from.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Repo folder for Project 7.
Lecture 8 — Generative & VLA Policies ≈2 hrs · Project 8 Expand ▾
  • 8.1 From policies to distributions ⚙️ In Progress
    Placeholder: from deterministic to probabilistic policies.
    Open Video ↗ Paste a URL later.
    Open Link ↗ IL notes / code examples.
  • 8.2 Diffusion policies intuition ⚙️ In Progress
    Placeholder: how diffusion models work for robotics policies.
    Open Video ↗ Paste a URL later.
    Open Link ↗ DAgger references + code.
  • 8.3 Vision-language-action models ⚙️ In Progress
    Placeholder: VLA models and multimodal robotics policies.
    Open Video ↗ Paste a URL later.
    Open Link ↗ RL primers + practice repos.
  • 8.4 Instruction following & grounding ⚙️ In Progress
    Placeholder: how robots understand and follow natural language instructions.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Offline RL references.
  • 8.5 Project 8 Overview ⚙️ In Progress
    Placeholder: project walkthrough and deliverables.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Sim2real + DR references.

Part V — Void: Learning → Humanoid

Lectures: 9–10 · Projects: 9–10 · Total: ≈4 hrs

“The question of whether machines can think is about as relevant as whether submarines can swim.”

— Edsger Dijkstra

Lectures & Units (Expandable)

Lecture 9 — Reflection & Meaning ≈2 hrs · Project 9 Expand ▾
  • 9.1 Why humanoids? ⚙️ In Progress
    Placeholder: why build humanoid robots and their unique advantages.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Embodiment essays, key papers, talks.
  • 9.2 Labor, automation, and power ⚙️ In Progress
    Placeholder: Labor, automation, and power dynamics in robotics.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Labor, automation, and power dynamics in robotics.
  • 9.3 Trust, safety, alignment ⚙️ In Progress
    Placeholder: building trustworthy and safe robotic systems.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Shared autonomy + interface refs.
  • 9.4 Sci-fi as design space ⚙️ In Progress
    Placeholder: using science fiction to explore robot futures.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Safety checklists, standards notes.
  • 9.5 Manifesto Overview ⚙️ In Progress
    Placeholder: project walkthrough and deliverables.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Ethics readings, policy notes, essays.
Lecture 10 — Final Integration ≈2 hrs · Project 10 Expand ▾
  • 10.1 System integration ⚙️ In Progress
    Placeholder: how the five parts connect into one coherent builder mindset.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Course map / summary notes.
  • 10.2 Debugging embodied systems ⚙️ In Progress
    Placeholder: debugging strategies for physical robot systems.
    Open Video ↗ Paste a URL later.
    Open Link ↗ License comparisons, standards docs, repo templates.
  • 10.3 Cost, reproducibility, openness ⚙️ In Progress
    Placeholder: making robotics accessible and reproducible.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Essays, sci-fi, philosophy readings.
  • Demo and evaluation criteria ⚙️ In Progress
    Placeholder: Demo and evaluation criteria.
    Open Video ↗ Prompt + examples talk.
    Open Link ↗ Template doc + exemplar projects.
  • 10.5 Final build overview ⚙️ In Progress
    Placeholder: project walkthrough and deliverables.
    Open Video ↗ Paste a URL later.
    Open Link ↗ Integration checklist / BOM template.