Note: This content is accessible to all versions of every browser. However, this browser does not seem to support current Web standards, preventing the display of our site's design details.

  

Distributed cooperative control and estimation for two KUKA Lightweight Robots

Student(en):

Betreuer:

Yvonne Stürz
Beschreibung:

This project contributes to the larger goal of controlling two KUKA Lightweight robots in a cooperative way through the realtime-interface FRI ("Fast Robot Interface").

In order to cooperatively manipulate objects, the robots are controlled in an impedance-control-mode, where they behave as a virtual mass-spring-damper-system. This flexibility helps to avoid high interaction forces and therefore possible damage to the object in the presence of uncertainties.

The main goal of the project is to implement a newly developed distributed estimation and control scheme on the KUKA LWR iiwa 14 R820 robots, which is based on underlying decentralized impedance controllers, available by KUKA.

To this end, the first step is to identify the model of the KUKA impedance-controller. Then, based on the identified parameters, the impedance controller is implemented in a simulation environment Gazebo, which has already been set up in a currently ongoing Master project. Based on the low-level decentralized impedance controllers, the distributed estimation and control scheme is implemented and tested in Gazebo.

After that, it is implemented on the KUKA LWR iiwa 14 R820 robots to be tested in small experiments and a demonstration of distributed cooperative manipulation of an object.

The project consists of the following steps:
  • familiarization with robotic dynamics (theory);
  • familiarization with the (real) robotic system i.e. perform simple tasks with the robot (Java-code library and C++ library (for the real-time-client) of KUKA are available);
  • familiarization with simulation environment Gazebo and ROS;
  • design and conduct experiments on the robot to identify the model parameters used in the impedance-control mode programmed by KUKA;
  • implement the KUKA impedance controller in Gazebo, using the identified parameters (ROS, C++);
  • implement the distributed estimation and control algorithms in Gazebo;
  • perform small experiments of cooperative manipulation on the robots to demonstrate the performance of the distributed control and estimation scheme.

Feel free to contact stuerzy@control.ee.ethz.ch for further information.



Weitere Informationen
Professor:

Roy Smith
Projektcharakteristik:

Typ:
Art der Arbeit:
20% Theory and Design
60% Implementation and Testing
20% Documentation
Voraussetzungen: Electrical or mechanical engineering background. Notions of robotics. Notions of dynamical systems, control and estimation design. Programming skills: C++, Matlab, Java. Experience with Gazebo and ROS.
Anzahl StudentInnen: 1
Status: taken
Projektstart: March 2016
Semester: FS 2016