Active vision for pose estimation applied to singularity avoidance in visual servoing

In active vision, the camera motion is controlled in order to improve a certain visual sensing strategy. In this paper, we formulate an active vision task function to improve pose estimation. This is done by defining an optimality metric on the Fisher Information Matrix. This task is then incorporated into a weighted multi-objective optimization framework. To test this approach, we apply it on the three image point visual servoing problem which has a degenerate configuration - a singularity cylinder. The simulation results show that the singular configurations of pose estimation are avoided during visual servoing. We then discuss the potential of active vision to be integrated into more complex multi-task frameworks.

I. Introduction and Background

Visual servoing is a motion control modality that closes the loop with visual information [1]. Because of this, it is very well suited to controlling the gaze direction of robots. For example, we have shown its application to humanoid robots in our previous works [2], [3]. The formulations in [2], [3] allows the gaze to be defined as a reference in image space.