In this chapter we investigate how to use knowledge representation and reasoning methods developed in artificial intelligence in order to implement robot control systems that can autonomously accomplish complex and underdetermined tasks, automatically adapt to task contexts, learn from experience, and forestall execution failures through action planning. We will use fetch and place tasks to be performed by a household assistant robot as our running example and show how these methods enable robots to make informed decisions about where to look for and find objects, how to pick them up, hold them, and where to place them.1 The chapter will introduce fact and rule-based knowledge representation as it is provided by logic and apply it to reason about robots, their components and capabilities, objects and environments, and actions and their effects. We will conclude by discussing how knowledge representation can be employed to equip robots with introspective reasoning capabilities, which allow the robot to answer queries regarding to what it is doing, why, how, what is predicted to happen, etc.