Dept. of Robotics & Automationat Luleå University of Technology

Master Thesis Projects in Robotics

The Robotics group at Luleå University is starting a research and industry project on controlling a mobile hydraulic arm using optronic feedback. Participants and sponsors include, in addition to the University, Holmhed Systems AB, Mobile Robotics AB, EU and Volvo Research Foundation. Two master thesis projects are proposed below. Other topics might also be possible, we are open for discussions.

The Robotics group is also looking for new PhD students. These master thesis projects might be a good starting point if you are interested in becoming a PhD student in robotics or finding a very interesting job in one of the participating high tech companies.

For more information contact:

Åke Wernersson.

Proposal 1:

Modelling, Identification and Control of a (Mobile) Hydraulic Arm.

Brokk 110The Brokk hydraulic arm is currently manually controlled. Your task is to design a computer control system for the arm, turning it into a heavy duty mobile robot arm.

Some problems to be solved:

  1. Derive a complete kinematic model for the robot.
  2. Derive a dynamic model for the arm, and identify/compute the relevant parameter values.
  3. Interface to the hydraulics - there is currently a radio link to the hydraulic actuators. You'll be able to use to that. However this will require programming and finding and buying the needed interface hardware.
  4. Model/identify the hydraulic components.
  5. Design a good "dead reckoning" model for the system. (I.e. predicting the state from the control inputs.)
  6. Design/test control laws. Optronic sensing for estimating the state of the system (in conjunction with your dead reckoning model) will be provided.
The project is appropiate for two, possibly three, students.

You can expect more support from your supervisors than is normal for a Master thesis project, and you will be cooperating both with the department and the participating companies. We expect you to take initiatives and appreciate any new ideas you can contribute.

A mix of different competences is very useful. Skills in control engineering and robotics are especially important. Programming and mechanics are useful. A desire to work with a combination of real hardware and mathemtical models is important.

The project will start as soon as possible, june is preferred.

Proposal 2:

Modelling of 3D-Range Measurements for Robot Control

RangescanCreating parametric models from 3D measurements is a complex and large problem area. This project will concentrate on models needed for realtime control of a robot performing specific tasks relative to an object. The object might be a steel bar to be grabbed or cut by the robot.

The number and complexity of objects/tasks will be decided by the supervisor and the students together. The project can also include robot control for actually executing the tasks, controlling the robot using the range sensor observations.

The project is appropiate for one or preferrably two students.

You can expect more support from your supervisors than is normal for a Master thesis project, and you will be cooperating both with the department and the participating companies. We expect you to take initiatives and appreciate any new ideas you can contribute.

A mix of different competences is very useful. Skills or interest in image processing, programming and robotics are especially important.

The project can start as soon as possible for you, summer or early fall is preferred.

Created by Johan Forsberg.
Last edited 2000-10-09 by Johan Forsberg.