Off-world Power Generation

Off-world Power Generation is a long-term project to develop autonomous in-space self-assembly: a fleet of identical robotic modules that find each other, maneuver, dock, and self-assemble a larger structure — the target application being self-assembling solar arrays at Sun–Earth L2. The hard, unsolved 90% of that problem is autonomy (relative navigation, docking, growth sequencing, fault recovery), and it does not need orbit to be developed — it needs a place where free-floating dynamics are cheap. This project builds that place on a desktop, publishes everything openly, and places its methods in the public domain as dated prior art.

Status (updated 2026-08)

• Air-bearing testbed built: 10-cm modules float on 6 mm float glass over a custom torsion-box bed, fine-leveled to 0.02° (target was <0.05°) — a planar (3-DOF) microgravity analogue.
• Simulation campaign complete: a closed-loop planar simulator plus perception, comms, and L2-navigation studies produced the findings behind the design — terminal docking precision is owned by the magnetic-capture envelope rather than the controller; multi-fiducial fusion is the dominant self-localization win; measured air/energy budgets per docking maneuver. The flight-portable autonomy code is the same code the testbed runs.
• Module v1 hardware received: Raspberry Pi Zero 2 W + wide camera (AprilTag relative pose), 8× Clippard solenoid valves (cold-gas thrusters), porous-graphite air bearings in 3D-printed housings, magnetic capture. Total project hardware spend to date: ≈ $1,500.
• Defensive publication live (2026-06-20): the autonomy, docking, slot-claiming, growth, and L2 relative-navigation methods are public-domain prior art — the design space stays open for everyone.
• Next milestone (Phase B): 10/10 repeatable autonomous docks between two modules on the testbed.

The labs

Air-bearing testbed

The shared apparatus: float-glass running surface, torsion-box bed, 3-point kinematic leveling, air-bearing modules. Underlies Labs 3–5.

Lab 5: Robotic self-assembly

The core autonomy work, including the defensive publication (2026) disclosing the project's in-space self-assembly autonomy, docking guidance, distributed slot-claiming, growth logic, and L2 relative-navigation methods as public-domain prior art.

Motivation: why a Nebraska-sized array

Premise

<Earth>   <-clearsky-burst-laser--   <satellite>

Goal

Near-earth or Lagrange-point solar array about the size of Nebraska to power the globe.

Calculations:

• Energy requirement of NYC: ~3000 trillion BTU in 2016 = 3000 trillion btu / 365 days = 3.4E+11 btu/hr = 1.0036680479e+11 watts = 100 GW
• PV power stations collect more power than solar thermal power stations. They seem to average ~3 MW/km^2 (throwing out 2 ridiculous outliers).
• Traditional single-junction cells have a maximum theoretical efficiency of 33.16% more. In reality it is around 18.7%.
• A 65"x39" (1.64 m^2) solar panel made in 2018 produces ~320W.
• About 48% of solar energy hitting the Earth reaches the surface. Perhaps optimistic, but we will divide by .48 to get energy-in-space vs on-earth.
• Approximate energy absorbable by a solar panel in space: 320W / 1.64m^2 / .48% = 400W/m^2 in space
• Approximate required size of a solar array "near Earth" (including l1) to power NYC: 100GW / 400W/m^s = 250 sq km
• 1 out of every 811 humans on earth live in NYC, so 250 sq km dish * 811 (assuming every human uses as much energy as a New Yorker) requires a solar array of only around 200,000 sq km - the size of Nebraska - to power the globe. A physicist confirms on the back of an envelope that the area needed will be "near-(UK)-country-sized".

The long road

Keep it as simple as possible, but no less.

• We need to optimize free space power transmission in lab conditions using currently-available consumer electronics.
• We need to create a solar panel array with robotics that can self-assemble. (← the current focus — see Status above)
• We need robotics that can precisely and safely aim laser energy to a distant target using a real-time handshaking protocol.
• We need to determine the cheapest possible way to launch a payload from Earth and navigate it to a final stable destination (lagrange or Earth orbit).
• We need to determine the requirements to receive laser power at an Earth-based power plant.