Robotics and Biology Laboratory

Effects of model complexity on speed and accuracy of visual servoing for manipulation

People

Georg Hieronimus

Clemens Eppner

Oliver Brock

Motivation

Manipulation tasks in robotics require high accuracy and precision in the position of its fingers. This is normally supported through visual data processing with a particle filter in order to achieve a recursive pose estimation. Nevertheless, the process might fail to provide the necessary speed. This work focuses on the estimation of the parameters that influence the accuracy and speed of the pose estimation. The number of particles being the most important one.

Description of work

This work approaches the Image Based Visual Servoing (IBVS) from three different technical aspects in order to improve the Speed and Accuracy of robotics grasping. Accuracy increase when tracking kinematic chains through the extension of the particle filter, speed up through GPU processing and strategic utilization of encoders in order to put the tracker on important parts of the input image.

 

Results

Manipulation tasks in robotics require high accuracy and precision in the position of its fingers. This is normally supported through visual data processing with a particle filter in order to achieve a recursive pose estimation. Nevertheless, the process might fail to provide the necessary speed. This work focuses on the estimation of the parameters that influence the accuracy and speed of the pose estimation. The number of particles being the most important one.