Supporting Semantic Capture during Kinesthetic Teaching of Collaborative Industrial Robots

Research output: Contribution to journalArticlepeer-review


Industrial robot systems being deployed today do not contain domain knowledge to aid robot operators in setup and operational use. To gather such knowledge in a robotic context requires mechanisms for entering and capturing semantic data. Such mechanisms would allow a system to gradually build a working vocabulary while interacting with the environment and operators, valuable for the bootstrapping system knowledge and ensuring the data collection over time. This paper presents a prototype user interface that assists the kinesthetic teaching mode of a collaborative industrial robot, allowing for the capture of semantic information while working with the robot in day-to-day use. Two modalities, graphical point-and-click and natural language, support capture of semantic context and the building of a working vocabulary of the environment while modifying or creating robot programs. A semantic capture experiment illustrates the approach.
Original languageEnglish
Pages (from-to)167-186
Number of pages20
JournalInternational Journal of Semantic Computing
Issue number1
Publication statusPublished - 2018 Apr

Subject classification (UKÄ)

  • Robotics


Dive into the research topics of 'Supporting Semantic Capture during Kinesthetic Teaching of Collaborative Industrial Robots'. Together they form a unique fingerprint.

Cite this