Overview : We developed NightHawk, a framework that combines active illumination with adaptive exposure control to optimize image quality in real time. NightHawk formulates an online Bayesian optimization problem to select the best light intensity and exposure time for each scene, using a novel feature detector–based metric as the optimization cost. Implemented as an event-triggered recursive optimization pipeline, NightHawk was deployed on a legged robot navigating a culvert beneath the Erie Canal.

🔍To see the image in full size, right-click on it.