Advanced

Optimal Tracking and Identification of Paths for Industrial Robots

Olofsson, Björn and Nilsson, Henrik (2010) In MSc Theses
Department of Automatic Control
Abstract
In many application areas in industrial production, industrial robots are utilised for performing various tasks. Frequently a predefined geometric path exists, such that the robot should track this path with its tool centre point. The tracking is often to be performed with certain criteria specified, such as minimisation of time or energy. Accordingly, path tracking problems can often conveniently be formulated as optimisation problems. This thesis concerns the problem of timeoptimal path tracking for industrial robots. The path tracking is experimentally evaluated on a robot from ABB of type IRB140 available in the Robotics Lab at the Department of Automatic Control, Lund University. In the thesis, mainly the optimisation software... (More)
In many application areas in industrial production, industrial robots are utilised for performing various tasks. Frequently a predefined geometric path exists, such that the robot should track this path with its tool centre point. The tracking is often to be performed with certain criteria specified, such as minimisation of time or energy. Accordingly, path tracking problems can often conveniently be formulated as optimisation problems. This thesis concerns the problem of timeoptimal path tracking for industrial robots. The path tracking is experimentally evaluated on a robot from ABB of type IRB140 available in the Robotics Lab at the Department of Automatic Control, Lund University. In the thesis, mainly the optimisation software JModelica.org has been used for optimisation purposes. In cases where the path only is defined by a motion of a tool along a contour of an object, experimental methods are required in order to determine the corresponding geometric motion of the robot. In the thesis a contact-force control approach for determining of the joint positions along the desired path is considered. Further, in a timeoptimal path tracking, one control signal is saturated in every time instance. Consequently, the robustness to modelling errors and disturbances is low. In order to make the control more robust, an earlier developed control structure called path velocity controller is implemented and tested in the robot system. Both contact-force controlled path identification and optimal path tracking are evaluated in simulations and in experiments on the robot system. (Less)
Please use this url to cite or link to this publication:
author
Olofsson, Björn and Nilsson, Henrik
supervisor
organization
year
type
H3 - Professional qualifications (4 Years - )
subject
publication/series
MSc Theses
report number
TFRT-5858
ISSN
0280-5316
language
English
id
8847449
date added to LUP
2016-03-16 12:40:40
date last changed
2016-03-16 12:40:40
@misc{8847449,
  abstract     = {In many application areas in industrial production, industrial robots are utilised for performing various tasks. Frequently a predefined geometric path exists, such that the robot should track this path with its tool centre point. The tracking is often to be performed with certain criteria specified, such as minimisation of time or energy. Accordingly, path tracking problems can often conveniently be formulated as optimisation problems. This thesis concerns the problem of timeoptimal path tracking for industrial robots. The path tracking is experimentally evaluated on a robot from ABB of type IRB140 available in the Robotics Lab at the Department of Automatic Control, Lund University. In the thesis, mainly the optimisation software JModelica.org has been used for optimisation purposes. In cases where the path only is defined by a motion of a tool along a contour of an object, experimental methods are required in order to determine the corresponding geometric motion of the robot. In the thesis a contact-force control approach for determining of the joint positions along the desired path is considered. Further, in a timeoptimal path tracking, one control signal is saturated in every time instance. Consequently, the robustness to modelling errors and disturbances is low. In order to make the control more robust, an earlier developed control structure called path velocity controller is implemented and tested in the robot system. Both contact-force controlled path identification and optimal path tracking are evaluated in simulations and in experiments on the robot system.},
  author       = {Olofsson, Björn and Nilsson, Henrik},
  issn         = {0280-5316},
  language     = {eng},
  note         = {Student Paper},
  series       = {MSc Theses},
  title        = {Optimal Tracking and Identification of Paths for Industrial Robots},
  year         = {2010},
}