An AI CubeSat (AISat) High Altitude Air Balloon Launching Mission for Atmospheric Research

During the past two decades, CubeSats have been developed dramatically for scientific and commercial use, with the advantages of their relatively low cost and short production time. We present here a novel system that can contribute to and expand observational capabilities in atmospheric and related sciences. For the past two years, the CubeSat-High-Air balloon Launch Mission (HAABSat) project has been successful in obtaining atmospheric data in the stratosphere (~101,000 ft). The CubeSat is assembled with only commercial off-the-shelf (COTS) components, with a frame that is 3-D printed. A suite of sensors that includes a barometer sensor, humidity particulate matter (PM) and CO2, measures the ambient environment. Additionally, an onboard gyroscope, accelerometer, and GPS are used for its navigational system. A magnetometer is placed 8 cm from the CubeSat body via a deployable boom to eliminate background electromagnet noise and collect accurate magnetic field measurements. The CubeSat consists of three onboard cameras, an ultra-wide-angle camera that takes high resolution (4056 x 3040 pixels) pictures, and a compact medium-wide lens with a 16 megapixels sensor that takes pictures of tele-objects, and a camera records videos at 1080P at 30 frames per second (FPS). Cameras were trained using AI to identify surface features of interest. Development of the CubeSat and AI training of cameras, data, and challenges of launch will be discussed.