As advanced robot systems steadily gain higher intelligence and greater autonomy, the requirements for the design of flexible interfaces to control these systems will increase. This is for instance the case, according to Rembold et al. , for the future use of intelligent robots in manufacturing or as service robots for different applications. Therefore, natural language as a communication medium for humans is an efficient means of making a technical system easier accessible to its users Wahlster . A practical advantage of natural language access is the possibility of conveying information in varying degrees of condensation and of communicating on different levels of abstraction in an application-specific way.
In this article, the joint efforts of the University of Karlsruhe and the University of Saarland to provide natural language access to the autonomous, mobile, two-arm robot, KAMRO (Karlsruhe Autonomous Mobile Robot), which is being developed at IPR, are reported. In the last two years, research has focused on the topic of task specification by use of natural language: operations and tasks which are to be performed by the robot can be specified at different levels. The most abstract specification would just provide a qualitative description of the desired state, i.e., the final positions of the different assembly parts. During task execution, the capability of recovering from error situations leads to a dynamic adjustment of the robot’s plans. This feature makes it more difficult for the operator to predict and understand the behavior of the robot. Explanations and descriptions of why and how a plan was changed would increase the cooperativeness of an intelligent robot. Thus, this article will focus on the extension of the implemented interface in allowing the use of natural language explanations to describe the intelligent robot’s error recovery.