An experimental study of predictors for visual servoing

In order to perform visual servoing tasks in a robotic system, one is confronted with the low sampling rate of standard cameras and the time delay introduced by image acquisition and processing. One way to circumvent the above problems is to estimate future positions of a moving object employing prediction techniques. In this work, three prediction techniques, namely Kalman filtering and two adaptive techniques employing least squares with forgetting factor and the projection algorithm respectively, are evaluated in terms of their prediction error and speed of convergence. Experimental results show that the adaptive technique employing the projection algorithm gives best results.