Intention recognition in manufacturing applications. (June 2015)

• Record Type: Journal Article
• Authors: Schlenoff, Craig; Kootbally, Zeid; Pietromartire, Anthony; Franaszek, Marek; Foufou, Sebti
• Is Part Of: Robotics and computer-integrated manufacturing. Volume 33(2015)
• Page Start: 29
• Page End: 41
• Publication Date: 2015-06
• Subjects: Intention recognition -- Manufacturing kitting -- Ontology -- State recognition -- Human–robot collaboration -- Robotics; Robots, Industrial -- Periodicals; Computer integrated manufacturing systems -- Periodicals; Robotics -- Periodicals; Robots industriels -- Périodiques; Productique -- Périodiques; Robotique -- Périodiques; 670.285
• DOI: 10.1016/j.rcim.2014.06.007
• Languages: English
• ISSNs: 0736-5845

Abstract

Abstract: In this article, we present a novel approach to intention recognition, based on the recognition and representation of state information in a cooperative human–robot environment. States are represented by a combination of spatial relations along with cardinal direction information. The output of the Intention Recognition Algorithms will allow a robot to help a human perform a perceived operation or, minimally, not cause an unsafe situation to occur. We compare the results of the Intention Recognition Algorithms to those of an experiment involving human subjects attempting to recognize the same intentions in a manufacturing kitting domain. In almost every case, results show that the Intention Recognition Algorithms performed as well, if not better, than a human performing the same activity. Highlights: A novel approach to intention recognition based on state recognition. Applied to the manufacturing kitting domain. States use a combination of spatial relations and cardinal directions. Algorithms performed better than humans performing the same task.