One of the important aspects of the robotic tasks is their variability. Most of the nowadays tasks are not pre-programmed and because of increased usage of reinforcement learning and other autonomous learning techniques, robotic trajectories have to be fitted in the robot workspace so that they have as much space for learning as possible, i.e., as high variability as possible. In this paper, we present and compare our methods for fitting a trajectory in a 2-D sagittal plane in the robot workspace. The method is shown on the example of robotic throwing action with the TALOS humanoid robot.
Fitting Constrained Trajectory with High Variability into Redundant Robot Workspace
Mechan. Machine Science
International Conference on Robotics in Alpe-Adria Danube Region ; 2022 ; Klagenfurt, Austria June 08, 2022 - June 10, 2022
2022-04-23
9 pages
Article/Chapter (Book)
Electronic Resource
English
Workspace fitting and control for a serial-robot motion simulator
Tema Archive | 2008
|A Robust Control Strategy of Redundant Manipulator by Workspace Observer
British Library Online Contents | 1996
|Representing the robot’s workspace through constrained manipulability analysis
British Library Online Contents | 2015
|