Online Calibration of One-dimensional Sensors for Robot Manipulation Tasks

The purpose of this paper is to enable a developer to easily employ external sensors emitting a onedimensional signal for flexible robot manipulation. To achieve this, the sensor must be calibrated using data tuples describing the relation between the positional change of the supervised object and the resulting sensor value. This information is used for adaptation methods, thus enabling robots to react flexibly to changes such as workspace variations or object drifts. We present a sensor-independent method to incrementally generate new data tuples describing this relation during multiple task executions. This method is based on the Secant method and is the only generally applicable solution to this problem. The method can be integrated easily into robot programs without detailed knowledge about its functionality. 2009 ICINCO – 6th International Conference on Informatics in Control, Automation and Robotics Milan / Italy