OpenSky · vehicle-compute network planner

OpenSky · vehicle-compute network

A fleet of vehicle-mounted Raspberry Pis harvest solar + wind while driving, run LLM inference + PoET proofs, and earn coins. The binding constraint is the per-unit energy budget, battery = health, running = drain, harvesting = capacity-limited replenishment, shutdown = stranding. The hierholzer_v scheduler plans the fleet; an economics layer turns scheduled uptime into coins, then revenue, then P&L, swept over device count N.

Levers

BOM solar panel: W 10 W deficit35 W surplus Fleet concurrency: % share of the fleet that can harvest at once Fleet size (solved): devices the engine actually schedules. Per-device energy is N-invariant, so the scale-out already projects to 1M; bigger just costs solve time (the gantt still draws 24 rows). Harvest window: h driving-in-sun hours per device Energy posture Revenue SKU Coin price: $ narrative coin to $ (no fiat price in the ledger)

Device fleet ·

Each row is a device over 180 days: wind + solar charge one battery, and it goes dark when that can't cover the load. online (idle) · computing (paid) · off · battery. Drag to zoom a range; double-click to reset.

Scale-out projection · the product number

The engine schedules a representative fleet; the economics extrapolate across device count N (harvest capacity scales with N, so per-device energy behavior is N-invariant). Watch the difficulty-damping bend past ~100 active devices, and whether the fleet pays back within its 36-month life.