Sprungmarken

Servicenavigation

Hauptnavigation

Sie sind hier:

Hauptinhalt

Detailseite



Trajectory Planning for Mobile Manipulators

Lehrstuhl: ET/IT RST

Betreuer: Christoph Rösmann, Frank Hoffmann,

Beginn ab: 19.10.2015

Maximale Anzahl der Teilnehmer: 6

Beschreibung: Robotic manipulation deals with the interaction of a robot with its environment. Usually, the interaction is realized with a tool attached to the robot arm and the concept itself is widespread in the industry, e.g. for welding and painting, stuffing printed circuit boards with IC components, inspecting and repairing structures and for simple pick and place tasks.


Due to increasing demands on complexity of the dynamic environment and motion tasks, robot arms cannot always operate from a fixed location, rather they must be able to combine navigation and interaction by means of extending the kinematic chain of the robot arm with a mobile base. Such a robot is called a mobile (wheeled) manipulator.
Attaching a robot arm on a mobile base requires new and extended control and planning schemes.
Especially in cases, in which the navigation and interaction part cannot be decoupled (due to constraints on the robot's kinematics or environment), the "complete" model of the robot must be integrated into the motion planning.

This project group consist of the development and evaluation of trajectory planners that explicitly preserve the coupling between arm and mobile base. Planners should be integrated into the framework of the "Robot Operating System" (ROS). ROS is a widespread C++/Python software library for robot applications and provides communication protocols, visualization utilities and a 3D simulator (to name just a few). In addition to simulation, real world experiments with a robot arm attached to a Pioneer 3DX mobile robot, demonstrating the advantage of taking coupling into account, are intended as part of the project group. The project group can refer to and build up on previous work in which an optimal trajectory planning algorithm is utilized for navigating the mobile robot base (without an attached arm). The work is available at http://wiki.ros.org/teb_local_planner.

Students are expected to have a background in robotics, control theory and optimization. Profound programming experience preferably in C/C++ is essential.