Something decides to act
The spacecraft’s onboard computer, or an AI planner, decides to do something: point a camera, fire a thruster, change mode, send data down.
How it works
Spacecraft are starting to make more of their own decisions, adjusting their own orbit, choosing what to photograph, responding to a problem without waiting for the ground. That’s valuable, but it also means a mistake can happen fast, with no one watching in real time. Vaelon is the independent check in between.
The spacecraft’s onboard computer, or an AI planner, decides to do something: point a camera, fire a thruster, change mode, send data down.
Before that action happens, Vaelon checks it against a fixed set of safety rules: is there enough power left, is it too hot or cold, is there enough fuel, is anything nearby, has anything failed. These rules don’t change and can’t be argued with.
If everything checks out, the action goes ahead. If not, Vaelon stops it and the spacecraft falls back to a safer action instead. Either way, the decision is written into a permanent record that can’t be edited afterwards.
Why it’s built this way
Think of it less like software and more like a mechanical interlock, the kind that stops a lift moving with the doors open. It doesn’t matter why the doors are open; the interlock doesn’t reason about it, it just holds. Vaelon works the same way: the safety rules hold regardless of what the onboard intelligence decides.
Vaelon Space
Make assurance part of the mission before autonomy becomes mission critical.