.png){kind=icon}
.png){kind=icon}
Return
Parking objects by pushing using uncalibrated visual servoing
.png){kind=icon}
PRE
AI
DOI：10.1007/s10514-018-9782-x.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Pushing is one of the strategies to perform robotic manipulation when the object is too large or too heavy. Motivated by this, we address the problem of how to push an object on a plane to a target pose with two cooperating robots. The main contribution is a new uncalibrated image-based control scheme that computes the required motion of the object to reach the target pose. Then, as an application of this control scheme, we study the conditions that allow performing the task of pushing the object with two robots. The setup consists of a fixed external uncalibrated camera looking at the workspace where the object and the robots stand. The task is defined with a target image of the object in the desired pose. The proposed control scheme computes the motion commands of the pusher robots and, as a result, they translate and rotate the object by imposing non-holonomic velocity constraints. This yields smooth, continuous and efficient trajectories. The stability of the control scheme is also proven. Experiments illustrate the performance of the control scheme.
Keywords:
Pushing
Manipulation
Visual servoing
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
4.3
Papers:
1.7K
Citations:
5.0K