Advances in deep space exploration via simulators & deep learning

The StarLight program conceptualizes fast interstellar travel via small wafer satellites (wafersats) that are propelled by directed energy. This process is wildly different from traditional space and trades large slow spacecraft for small, fast, inexpensive, fragile ones. main goal of these to gather useful images during their deep journey. We introduce solve some the problems accompany this concept. First, we need an object detection system can detect planets have never seen before, containing features may not even know exist in universe. Second, once exoplanets, a way take rank them importance. Equipment fails data rates slow, thus method ensure most important humankind ones prioritized transfer. Finally, energy on board minimal must be conserved used sparingly. No exoplanet should missed, but using erroneously would detrimental. simulator-based methods leverage artificial intelligence, mostly form computer vision, order all three issues. Our results confirm simulators provide extremely rich training environment surpasses real images, train models yet observed humans. also show immersive adaptable provided simulator, combined with learning, lets us navigate save otherwise implausible way.