Collaborative programming of robotic task decisions and recovery behaviors

Abstract Programming by demonstration is reaching industrial applications, which allows non-experts to teach new tasks without manual code writing. However, a certain level of complexity, such as online decision making or the definition recovery behaviors, still requires experts that use conventional programming methods. Even though, cannot foresee all possible faults in robotic application. To encounter this, we present framework where user and robot collaboratively program task involves behaviors. Hereby, task-graph created represents production alternative Nodes represent start, end states links define actions for execution. This graph can be incrementally extended autonomous anomaly detection, requests add knowledge specific action. Besides our proposed approach, introduce two approaches manage behavior compare extensively study involving 21 subjects. revealed strength analyzed how users act robot. Our findings proclaim with based representation detection not only initiating but particularly transfer those