Visual servoing

ViSP provides visual features and control laws that allow building visual servoing applications in order to control the motion of a robotic system. Below you will find videos that show very simple examples of visual-servoing applications.

Robot positioning wrt an object

The first three videos show the behavior of an IBVS, a 2D 1/2 and a PBVS visual-servoing scheme during a positioning task achieved on an ADEPT Viper s850 arm.

This video shows the behavior of an image based visual servoing (IBVS) scheme in order to position the robot with respect to a simple target. In the image, the trajectories of the points are straight lines. [source code]		This video shows the behavior of a 2D half visual servoing (2D 1/2) scheme. In the image, the trajectory of the reference point is a straight line. [source code]

This video shows the bahavior of a position-based visual servoing (PBVS) scheme. The 3D trajectory of the camera frame is a straight line. [source code]

These two videos show how it is possible with ViSP to build a visual servoing scheme from four line features in order to position a robot with respect to a postcard. On the left the camera is fixed on an Afma robotics gantry robot. On the right the camera is fixed on an ADEPT Viper s850 six dof arm. After convergence, the target is moved by hand. [source code]

Target tracking

This example shows how it is possible with ViSP to build a visual servoing target tracking scheme from an ellipse visual feature in order to control the pan and tilt of a camera to keep the ball centered in the image. The zoom of the camera is also controlled in order to keep the surface of the ball constant in the image. [source code]

Joint-limits avoidance

This video shows an example of coupling visual servoing and joint-limits avoidance. The control scheme combines the regulation at zero of the selected vision-based task to gaze a target with the minimization of a secondary cost function. This cost function reflect the manipulability of the robot in the vicinity of the joint limits. The [source code] is an implementation of the paper E. Marchand, F. Chaumette, A. Rizzo. Using the task function approach to avoid robot joint limits and kinematic singularities in visual servoing. In IEEE/RSJ Int. Conf. on Intelligent Robots and Systems, IROS'96, Volume 3, Pages 1083-1090, Osaka, Japan, November 1996.