Robots never know exactly where they are, what they see, or what they're doing. They live in dynamic environments, and must coexist with other, sometimes adversarial agents. Robots are nonlinear systems that can be underactuated, redundant, or constrained, giving rise to complicated problems in automatic control. Many of even the most fundamental computational problems in robotics are provably hard. Over the years, these are the issues that have driven my group's research in robotics. Topics of our research include visual servo control, planning with uncertainty, pursuit-evasion games, as well as mainstream problems from path planning and computer vision.