Towards Automatic Error Recovery in Robot Programs

Unexpected events can cause the failure of apparently "correct" robot programs. The interaction with the real world and its unpredictability make the problem of error recovery in robot programming specially important. The goal of the paper is to present a general framework in which the activity of error recovery can be automated. This is accomplished by introducing a monitor program which identifies the appearance of any error and attempt to correct that error. The correction is done using a knowledge base where the knowledge that the user has about error identification and correction is expressed in symbolic form. An inference mechanism allows extension of this knowledge base for use in complex and unanticipated situations.