Free estimate
Menu

Case studies

Five days of work: a satellite sensor proven in the wilderness

From a goal with no spec to a working satellite sensor in 5 days of work, on under $1,300 of off-the-shelf hardware, tested hours into the wilderness. Its frames hold real satellite tracks, identified by name.

Our own research prototype, built from off-the-shelf parts. We designed, built, and fielded it.

5 days of workfrom ideation to a sensor in the field
Under $1,300of off-the-shelf hardware

Starting point: a goal with no spec and a remote test site

The goal was a low-cost sensor that sees satellites, built from parts anyone can buy, with no spec. We built it alongside our customer work: about five days of full-time effort, spread across under six weeks. For the field test we chose the Gila National Forest near Quemado Lake, New Mexico, for its exceptionally dark skies.

Approach: design the system, test in the field, analyze the data

  1. Design a model (CAD).
  2. Build the prototype from off-the-shelf parts. A monochrome camera with a lens of about 16 mm sat on the mount we designed, which adapts to a tripod, and recorded one frame every second to a laptop.
  3. Field it for three nights to collect data.
  4. Stack frames aligned on the stars, then match the satellite tracks. The stack shows each satellite’s path across the frames, and we matched each path to published pass predictions to identify the satellite by name.

We modeled the system in SysML: the model shows any customer how the sensor works and how it connects to their systems, as one of a family of sensors at the edge.

The parts, the field settings, and the processing are in A staring satellite sensor from off‑the‑shelf parts: field results.

Three real artifacts in order: our design model (CAD render) of camera, lens, arm mount, and edge computer, made before the field trip; the fielded prototype, a camera on a tripod capturing to a laptop; and its 15 stacked frames with two satellite tracks.1 Design modellenscameraedge computerarm mount2 Fielded prototypecameralaptop3 Stacked framesDesign model (CAD render), made before the field trip: camera, lens, arm mount, and edge computer.1 Design modellenscameraedge computerarm mountThe fielded prototype: the camera on a tripod, capturing to a laptop.2 Fielded prototypecameralaptop15 one-second frames, stacked: two satellite tracks.3 Stacked frames

Results: satellites identified by name

Hours into the wilderness, the sensor recorded real images and real satellite tracks; a couple of hours of analysis named satellites, and the frames hold far more tracks than we pulled out by hand.

The same crop as one frame and as 15 frames stacked: each satellite is a short dash in the single frame and a full track in the stack; the same short marks sit beside the same places.1 frame15 frames, stackedThe same crop as one frame and as 15 frames stacked: short dashes in the single frame, full tracks in the stack.1 frame15 frames, stacked
Fig. 1 Stacking 15 one-second frames turns faint dashes into full satellite tracks.
Chart of the signal-to-noise gain from stacking, the square root of the number of frames: 4 frames give 2 times; our 15 frames give about 3.9 times.signal-to-noise gain1×2×3×4×14916frames stacked (N)√N4 frames: 2×our 15 frames:≈3.9×
Fig. 2 Stacking raises the stars’ signal-to-noise with the square root of the number of frames; each satellite’s track shows across the whole stack.

Engineering write-up: A staring satellite sensor from off‑the‑shelf parts: field results

Buy from us or team with us: Contracting

Start from the goal; no spec needed.

Estimates are free. Send the scope through the contact form; we reply within one business day and send a written quote within five business days.

Get a free estimate