Systemsthat watchthe sky
In February 2009, two satellites were tracked as a safe miss — half a kilometre apart. They collided anyway, at 11.7 km/s, scattering thousands of fragments across low Earth orbit. The tools that missed it measured distance. I build ORCAS: software that measures probability instead, and would have flagged that collision as critical.
A “safe distance” isn’t always safe.
Every tracked satellite position carries some uncertainty — less a point than a small cloud of possible locations. Two satellites can be reported half a kilometre apart and still have those clouds overlap enough to collide. Distance alone can’t see that; only the shape of the uncertainty can.
ORCAS combines classical orbital mechanics with machine learning to compute that risk directly — a real probability of collision (Pc), for thousands of objects at once, not just a yes/no distance check.
They weren’t wrong about the distance in 2009. They were wrong to treat it as a distance.
Background & Path
B.Tech Computer Engineering · Silver Oak University
Started B.Tech
Computer Engineering, Silver Oak University, Ahmedabad.
Quantum-Enhanced AES
Course research project — quantum-random key generation via Qiskit, validated against NIST SP 800-22.
Team Lead — Space Debris Tracking
Leading a multi-member team building orbital debris tracking, prediction and visualisation on NASA APIs and public satellite tracking data (TLEs).
Paper accepted
ICSSIT 2026, Paper ID 1849, technically sponsored by the IEEE SMC Society. Presented 28–30 July.
Paper published
Indexed in IEEE Xplore. ICSSIT 2026 proceedings, pp. 1957–1962.
Graduating
B.Tech complete. Targeting research groups working on space situational awareness.
ORCAS Deep Dive
Systems & Research
ORCAS
Python · AstrodynamicsPredicts whether two satellites will actually collide — not just how close they'll pass — and reconstructs real historical near-misses to prove it. Began as the Space Debris Tracking System team project at Silver Oak University.
Quantum-Enhanced AES
Qiskit · CryptographyEvaluating quantum-random key generation via Qiskit against classical PRNGs with NIST SP 800-22 statistical validation.
Diablex Core
Node.js · React · FHIRMedical data analytics platform processing 1,000+ daily continuous glucose measurements with FHIR API interoperability.