We work on the mathematics that keeps aircraft apart.
GodwitNav Solutions is a small engineering practice in Falls Church, Virginia. We derive closed-form methods for detecting and resolving losses of separation between aircraft, publish them with their validation, and offer independent technical verification of other teams’ separation and deconfliction algorithms.
Three lines of work, one set of methods
The research produces the methods. The verification service applies the same discipline to other people’s algorithms. The autonomy work carries the methods toward drones and flying hardware.
- Complete
- Submitted
- In progress
- Planned
-
Conflict detection and resolution research
Closed-form solutions for predicted losses of horizontal separation between two level aircraft, with first-order propagation of measurement uncertainty into every resolution variable. Validated against direct trajectory simulation and nonlinear Monte Carlo.
Preprint published
Journal manuscript
-
Independent technical verification
We reconstruct what an aviation algorithm mathematically means, build an independent reference implementation, and produce rerunnable evidence of where it matches its requirements and where it does not. Aimed at UTM, advanced air mobility, and autonomy teams.
Taking first engagements
-
Autonomous UAS platform
An ArduPilot and NVIDIA Jetson research drone designed so flight control stays independent of the companion computer, built to carry our methods onto real aircraft.
Simulation & CI
First airframe
Civil and defense
The same questions matter in defense: where an uncrewed vehicle will be, how sure that prediction is, and whether the software acting on it can be trusted. We offer counter-UAS, autonomy test and evaluation, and trajectory and intercept analysis to Army and DoD programs.
What we can point to
Our own conflict-resolution paper went through the same kind of checking we offer customers. The checking found a real problem, which is the point of doing it.
Preprint and validation supplement (DOI 10.5281/zenodo.21956633)
- 372
- comparisons in an audit of a draft table revision. The audit isolated a systematic sign defect in the draft uncertainty totals.
- 28
- automated analytical and direct-simulation tests, passing on the frozen audit environment.
- 15,000
- randomized encounter geometries checked by brute force at 0.25° heading resolution, with no misclassification.
- 144
- published Monte Carlo runs (48 cases × 3 seeds × 200,000 samples), regenerated byte-for-byte from the archived script, seeds, and configuration.
Who we are
GodwitNav Solutions LLC was founded by Aydin Jalilov, founder and principal engineer, who leads the company’s software, verification, and autonomy work and brings over fifteen years in airspace design and aviation safety analysis (about the founder). The conflict-resolution research is coauthored with Firdovsi Jalilov, an independent researcher; both authors contributed jointly to its methodology, validation, and manuscript.
GodwitNav is named for the bar-tailed godwit, which flies the longest non-stop migration of any bird. Over open ocean it cannot land to correct a mistake, so its navigation has to be right from the start. It doesn’t fly alone. In a flock, every bird keeps asking whether it is about to come too close to another, and which way to turn. Flocks turn together without colliding. Our work sits in those same two questions: navigating with no room for error, and keeping aircraft apart.
Get in touch
For a verification conversation, a question about the research, or a teaming inquiry, email info@godwitnavsolutions.com. Tell us what the algorithm does and what decision the evidence needs to support.
Registration
- UEI
- FX6FGLFH7XQ5
- CAGE
- 237W6
- SAM.gov
- Active
- Size
- Small business