Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Modelling and Control of a Parallel Kinematic Robot

Rosquist, Kristofer (2013)
Department of Automatic Control
Abstract
This Master’s Thesis shows how to model the kinematics and forward dynamics for the parallel kinematic robot from ABB, IRB340 FlexPicker, and how to implement the control using software and hardware from B&R Automation. With conventional methods the forward kinematics, inverse kinematics and the kinematics for the velocities and accelerations of the robot are modelled. The forward dynamics, calculating the generated motor torques, of the robot is modelled with MapleSim and code generation of this model, to be implemented in the software from B&R
Automation and used online during control, is used. In the software from B&R Automation, B&R Automation Studio, the kinematic model, path trajectories for the TCP with polynomial functions, an... (More)
This Master’s Thesis shows how to model the kinematics and forward dynamics for the parallel kinematic robot from ABB, IRB340 FlexPicker, and how to implement the control using software and hardware from B&R Automation. With conventional methods the forward kinematics, inverse kinematics and the kinematics for the velocities and accelerations of the robot are modelled. The forward dynamics, calculating the generated motor torques, of the robot is modelled with MapleSim and code generation of this model, to be implemented in the software from B&R
Automation and used online during control, is used. In the software from B&R Automation, B&R Automation Studio, the kinematic model, path trajectories for the TCP with polynomial functions, an interface between the software and a PC and programs for communicating with the drives are also implemented together with a main program. To simplify, analyze and calculate equations Maple is used.

Experiments using the generated torque are made for comparison with the model and to improve it. The model is improved by measuring the friction torque and the torque caused by the inertia of the motor, a gear and one robot arm.

Because of the parallel structure of the robot the generated dynamic equation consists of complex differential equations which can overload the PLC during solving. To ensure stability of the dynamic model a second dynamic model with constraint stabilization is generated for use during slower movements. Despite the risk of starvation of the PLC the robot is moved at 10 g acceleration with good result comparing calculated and actual torque. Without using the desired feedforward from the dynamic equations, but instead using a simpler less dynamic feedforward method, the robot is accelerated at 16 g. (Less)
Please use this url to cite or link to this publication:
author
Rosquist, Kristofer
supervisor
organization
year
type
H3 - Professional qualifications (4 Years - )
subject
ISSN
0280-5316
other publication id
ISRN LUTFD2/TFRT--5929--SE
language
English
id
4332966
date added to LUP
2014-03-03 11:28:32
date last changed
2014-03-03 11:28:32
@misc{4332966,
  abstract     = {This Master’s Thesis shows how to model the kinematics and forward dynamics for the parallel kinematic robot from ABB, IRB340 FlexPicker, and how to implement the control using software and hardware from B&R Automation. With conventional methods the forward kinematics, inverse kinematics and the kinematics for the velocities and accelerations of the robot are modelled. The forward dynamics, calculating the generated motor torques, of the robot is modelled with MapleSim and code generation of this model, to be implemented in the software from B&R
Automation and used online during control, is used. In the software from B&R Automation, B&R Automation Studio, the kinematic model, path trajectories for the TCP with polynomial functions, an interface between the software and a PC and programs for communicating with the drives are also implemented together with a main program. To simplify, analyze and calculate equations Maple is used.

Experiments using the generated torque are made for comparison with the model and to improve it. The model is improved by measuring the friction torque and the torque caused by the inertia of the motor, a gear and one robot arm.

Because of the parallel structure of the robot the generated dynamic equation consists of complex differential equations which can overload the PLC during solving. To ensure stability of the dynamic model a second dynamic model with constraint stabilization is generated for use during slower movements. Despite the risk of starvation of the PLC the robot is moved at 10 g acceleration with good result comparing calculated and actual torque. Without using the desired feedforward from the dynamic equations, but instead using a simpler less dynamic feedforward method, the robot is accelerated at 16 g.},
  author       = {Rosquist, Kristofer},
  issn         = {0280-5316},
  language     = {eng},
  note         = {Student Paper},
  title        = {Modelling and Control of a Parallel Kinematic Robot},
  year         = {2013},
}