A Robot Control System Integrating Reactive Control, Reasoning, and Execution Monitoring

We present a robot control system that integrates robust reactive control with efficient reasoning about actions and execution monitoring. On the reactive level, the robot is controlled using a hierarchy of low-level behaviors. On the high level, a logical representation of the world enables the robot to reason about the state of the world and to plan action sequences. If the execution of an action fails, probabilistic information on the state of the world based on the robot’s sensors are used to update the logic-based representation of the world. High-level reasoning then allows to infer possible explanations and to recover from the failure situation. The proposed system is evaluated using a set of realistic office-delivery scenarios.