Capture low-resolution Earth imagery and downlink photos via UHF to a student-built ground station. Use images for outreach, attitude verification, and as training data for future on-board image classification experiments.
Capture low-resolution Earth imagery and downlink photos via UHF to a student-built ground station. Use images for outreach, attitude verification, and as training data for future on-board image classification experiments.
This is a beginner-level project with an estimated timeline of 8-12 months using a 0.5U form factor.
There is something uniquely compelling about receiving a photograph taken by a satellite you built yourself. A low-resolution Earth imaging camera is arguably the highest-impact outreach payload available the images may not rival commercial constellations, but they are yours. The camera captures visible-light photographs of Earth's surface and atmosphere, stores them onboard, and downlinks them during ground station passes. From a typical low Earth orbit altitude, even a modest two-megapixel sensor produces recognizable images showing coastlines, cloud formations, city grids at night, and large geographic features. Beyond outreach, the imagery serves practical engineering purposes: verifying that the satellite is pointing where it should be, assessing attitude stability from frame-to-frame consistency, and generating training data for future onboard image classification experiments. The project covers optics mounting and alignment, image compression and storage, downlink scheduling and bandwidth management, and ground-side image decoding and georeferencing. For teams building their first satellite, this payload provides immediate, visual confirmation that the entire system structure, power, comms, and pointing is working together.
ArduCAM OV2640 (2MP, ~$15-25, SPI for data / I2C for control) produces recognizable Earth images at ~100-500 m/pixel GSD from 400-500 km. Mount on nadir face of payload module with clear optical window in Al-6061 enclosure. Store JPEG images on microSD via SPI. Implement simple scheduling: capture every N minutes when sunlit, skip eclipse. Ground software: Python script to decode, geolocate (using orbital TLE + timestamp), and mosaic images. Consider OV5640 (5MP, ~$30) for slightly better resolution.
CMU PyCubed-Mini includes camera module natively direct reference implementation available. Utah State GASPACS (2022) used Raspberry Pi Zero + Pi Camera in 1U. Resolution will not compete with commercial imagers but "the first photo from your own satellite" is an extraordinarily powerful recruitment and outreach tool. I2C bandwidth bottlenecks image transfer use SPI for data. 2MP JPEG compresses to ~200-400 KB, requiring ~8-16 seconds over SPI. Camera costs $100-$500 total and fits trivially in 0.5U. Complexity: low-to-medium. Perhaps the most engaging option for students and easiest to integrate with PyCubed. Can serve as input data source for project 4 (image classification) or project 23 (image triage).
This project spans 2 disciplines, making it suitable for interdisciplinary student teams.
Ready to take on this project? Here's a general roadmap that applies to most CubeSat missions:
Connect with a Blackwing chapter for mentorship, platform access, and a path to orbit.