// PROJECTS · 29 WEEKS

Ten projects from a home lab.

Start on the laptop, then move to the arm on the desk, then to a rover in the garage. Projects 1–7 are the spine. Projects 8–10 are optional depth if the core is solid. Each one ends with a folder you can open, a short clip, and a row in the shared eval log.

IN PROGRESSP01 · Beginner · core

MuJoCo foundations

Physics and control on the laptop: hand-written PD, a scripted fish gait, then a small PPO swimmer. Everything lands in the harness.

Weeks 1–2 · Sim only · MacBook · $0 · 01_mujoco_foundations/

MuJoCo 3.x (native Apple Silicon)Train · MacBook M4 MaxRun · MacBook M4 MaxBuy · $0ROS · NoGazebo · No
MuJoCo foundations
IN PROGRESS
DEMO 01 · IN PROGRESS
DEMO 01 · IN PROGRESSVideo lands here when this milestone ships.
Fish swimming under the scripted CPG gait and under PPO, plus the held-out results table.
Parallel bodies in a physics sim
Parallel bodies in a physics sim
Training floor for later projects
Training floor for later projects
MuJoCoPD controlCPGPPOEval Harness
01

Goal

Load a stock MJCF model, step physics, and open the viewer. Learn the XML: bodies, joints, actuators, geoms, contacts. Write PD by hand on a pendulum (do not import a controller). Tune until it holds a setpoint, then put the same gains on a double pendulum and write why it fails. Write down the observation vector, action units, and loop rate. Build a 3–5 link robotic fish with a CPG gait on MuJoCo hydrodynamics. Learn the same behaviour with a small PPO run toward randomized targets. Compare scripted vs learned on the same held-out set. Change one reward term. Ship common/eval_harness/ v0 with run_type already accepting sim | real | paired.

02

Why it matters

This lab repeats one pattern: model something, write a controller, test on data you held out, log it, then fix what was wrong. Project 1 does that once on a laptop so the pieces are concrete before anything expensive depends on them. Cartpole as the whole project is out. A balanced pole convinces nobody. The fish stays in sim on purpose. Waterproofing a printed fish is a month of sealing and servo failures with almost no sensing, and there would be no sim-to-real gap to measure. Put “real” on the harness schema now or you re-instrument everything later.

03

The work

  1. 01

    Load a stock MJCF model, step physics, render the viewer. Learn bodies, joints, actuators, geoms, contacts.

  2. 02

    Write PD by hand on a pendulum. Tune until it holds a setpoint. Carry the same gains to a double pendulum and write why it fails.

  3. 03

    Write down observation contents, action units, and loop rate. This is the most common later bug source.

  4. 04

    Build a 3–5 link swimmer with a CPG gait (sinusoidal joints, phase-offset down the body, PD-tracked) on MuJoCo hydrodynamics.

  5. 05

    Small PPO run toward randomized targets. Compare against the scripted gait on the same 20 held-out targets. Change one reward term; capture before/after.

  6. 06

    Ship common/eval_harness/ v0 logging every run, with run_type: sim | real | paired.

04

Done looks like

  • p1/SETUP.md with exact package versions and a one-command repro script.
  • A video of the fish swimming under the scripted CPG gait.
  • A video of the fish swimming under the learned PPO policy.
  • Results table: scripted vs learned on 20 held-out targets (distance-to-target, time-to-reach, path efficiency).
  • One reward-term change with a before/after behaviour clip and the numbers.
  • common/eval_harness/ v0 logging every run above, run_type already accepting sim | real | paired.
05

Stack card

SimulatorMuJoCo 3.x (native Apple Silicon)
Trains onMacBook M4 Max
Runs onMacBook M4 Max
Buy$0
ROS 2No
GazeboNo
Compute / targetNone
Installmujoco, mujoco-python-viewer, numpy, matplotlib, gymnasium, stable-baselines3, Python 3.11, uv or conda
06

Repos to study

07

Watch out

  • On macOS the viewer must be launched with mjpython / make view, never plain python.
  • Do not overwrite configs/heldout_targets.json. Freeze the set before training.
  • The fish stays in sim. Do not spend a month waterproofing a printed fish.
08

Workstreams

BUILD STREAM

MJCF models, hand-written PD, CPG gait, PPO training on the laptop, SETUP.md and one-command repro.

EVAL STREAM

Frozen 20-target set, scripted-vs-learned table, reward A/B, harness v0 with sim | real | paired.

Cross-check. Either stream can launch make smoke / make eval and read the other's numbers.

09

What this proves

You can read a simulator’s API, write a controller instead of importing one, and design a held-out experiment before anyone hands you a GPU.

Read a simulator’s API, write a controller instead of importing one, and design a held-out experiment before anyone hands over a GPU.

10

Study videos

11

Gear / BOM

MacBook M4 Max
MacBook M4 MaxMuJoCo, LeRobot recording, control station. Not the Isaac engine
ownedIN PROGRESS
Gears: hardware + stack →