DISAMBIGUATION (renamed 13 Jun 2026 from
lunar-teleoperation): this page is about operating robots on the physical Moon despite Earth-Moon latency. It is NOT about Lunar Ventures. For the Lunar Ventures / Elad Verbin Earth-robotics teleoperation INVESTMENT thesis see Teleoperation Bridge; for the autonomy destination see Robot Autonomy Destination. The old sluglunar-teleoperationnow redirects here.
Active research thread. Half-day per week. Connected to State of the Future writing + Telekine DD work.
What it is
Operating robots and equipment on the lunar surface from Earth despite the Earth-Moon round-trip latency (~1.3-2.7 seconds). Distinct from full autonomy (where the on-board AI decides) and from low-latency teleoperation (where humans drive directly). It’s a hybrid: human strategic intent + AI tactical execution + delayed-feedback human-in-the-loop oversight.
Why now
Three converging signals:
- NASA Artemis + commercial lunar landers create a market for surface operations from 2026-2030.
- CHIPS Act + space industrial policy funds rad-hard / lunar-grade compute and sensors.
- Foundation models + closed-loop control are bridging the human-in-the-loop / autonomous gap on Earth (Wayve, Tesla FSD lineage). The same architectures may transfer to lunar with latency adapters.
Open research questions
| Question | Notes |
|---|---|
| Latency budget | What closed-loop control is feasible at 2.6s round-trip? Predictive layers, on-board safety reflex, delayed human approval — the architecture that handles all three. |
| Hardware stack | Lunar surface: ±150°C thermal cycle, vacuum, radiation, dust. What compute/comms hardware survives? |
| Operator UX | How does a human pilot through latency? Tesla / Wayve patterns vs. surgical robotics vs. drone teleop. |
| Mission economics | Hours of operator time vs. autonomy hours — what’s the human-cost-per-task crossover? |
| Comms architecture | Direct-to-Earth vs. lunar-orbit-relay vs. mesh networks |
Active threads
- Telekine DD work — Lunar Ventures pipeline; specific company doing teleoperation for industrial/extreme environments. (Tracked in the Earth-robotics thesis Teleoperation Bridge.)
- Stan Boland conversations — Wayve founder, autonomous systems lineage. Relevance: foundation-model-based driving stacks may port to teleop.
- Wayve latency adaptation — open question, worth a direct conversation.
- Global Foundries lunar-grade silicon — does GF have a rad-hard process? Worth asking the LP.
Connection to investing
Sources
Frontier
See open research questions table above.