A full conjunction-assessment pipeline, delivered as software.
No ground stations to build, no analysts to staff overnight. Photon runs the entire loop — ingest, analyze, alert, maneuver — and hands the team clear decisions.
One trusted catalog from many noisy sources
Photon fuses public TLE/OMM data with commercial radar feeds, deconflicts duplicates, and continuously reconciles state estimates — so decisions are never made from stale or contradictory orbits.
ML that tightens the math and cuts the noise
Learned models refine covariance, anticipate maneuvering objects, and rank every encounter by true probability of collision — so the team chases real threats, not artifacts of bad data.
The right warning, where the team already lives
Deduplicated, severity-ranked alerts arrive by Slack, email, webhook or API — with the full encounter context attached and escalation only when probability crosses a threshold.
The smallest burn that clears the threat
Photon searches the maneuver space for the most fuel-efficient option that drives collision probability below the limit, simulates the outcome, and hands over a plan to review and command.
Built for real orbits, honest about the rest
A Rust core does the number-crunching; Python runs the AI and the 3D console. Everything is screening-level and traceable — Photon proposes, an operator decides.
Screen the fleet against the full public catalog — payloads, rocket bodies and debris.
Propagation and screening run on the operator's machine and keep working without a connection.
Runs on public tracking data and feeds already in place. Nothing to install in orbit.
Every suggestion logs its data, prompt and reasoning — conservative, screening-level, reviewable.
See Photon on your own fleet
Bring the catalog IDs and Photon will show live conjunctions in a guided demo.
Book a demo