Iterative learning control based on sensor fusion

Photo: ABB

The aim of the research is to improve the accuracy of industrial robots by using Iterative Learning Control in based on estimates of signals obtained by sensor fusion.

Iterative Learning Control makes use of the fact that industrial robots often carries out operations repeatedly, and that the algorithms that determine the input signals to the robot iteratively can learn and improve the way to control the robot. In order to achieve this it is necessary to measure or estimate signals that give information about e.g. the position and orientation of the tool that is mounted on the robot.

One research problem is to study how the information from the built in sensors in the robot can be combined (fused) with the information from other sensors and make use of the knowledge of the properties of the robot in order to get an as accurate as possible estimate of the relevant signals.

The research is carried out within the VINNOVA-supported Industry Excellence Center LINK-SIC and in collaboration with ABB Robotics.

Contact

Staff at Automatic Control

About the division

About the department