Model-based reasoning of sensory data for robotic assembly

Research output: Contribution to journalArticlepeer-review

Abstract

Sensory data provide crucial information for controlling and monitoring the execution of an assembly operation. Moreover, in the event that an error occurs during the assembly operation, this information is needed to determine the state of the assembly operation, the type of error, and the appropriate error-recovery procedure to correct the error. This paper addresses the problem of monitoring and interpreting sensory data acquired during robotic assembly. Sensory data are classified as spatial, physical, and temporal, and abstract model for each type of sensory data are developed. Each sensor provides information about the state of the assembly operation and through the use of evidential reasoning, this information is combined to obtain a combined state of the assembly operation. With this combined state of the assembly operation, we can monitor the progress of the assembly operation. In the event of an error in the assembly operation, error-recovery procedure based on low-level motions of the manipulator are selected depending on the sensory information.

Original languageEnglish (US)
Pages (from-to)490-494
Number of pages5
JournalProceedings of SPIE - The International Society for Optical Engineering
Volume1198
DOIs
StatePublished - Mar 1 1990

All Science Journal Classification (ASJC) codes

  • Electronic, Optical and Magnetic Materials
  • Condensed Matter Physics
  • Computer Science Applications
  • Applied Mathematics
  • Electrical and Electronic Engineering

Fingerprint Dive into the research topics of 'Model-based reasoning of sensory data for robotic assembly'. Together they form a unique fingerprint.

Cite this