My research is focused on developing robust, multisensor perception algorithms that enable robots to operate safely over long periods of time and in challenging environments, for example, on road networks, underground, underwater, in space, and on remote planetary surfaces. My goal is to create highly independent autonomous robotic systems for a wide range of tasks, from exploring the surface of Mars to helping with eldercare here on Earth. A central concept in my work is the notion of introspection and internal model maintenance: successful long-term autonomy will require robots to maintain accurate representations both of the external world and of the state of their own sensors and actuators over time. More generally, I am interested in 3D computer vision, probabilistic modeling, estimation theory, and machine learning. I've also recently developed an interest in cloud robotics.