Visual servoing is generally based upon geometrical features. Recent developments were made in the way of a generalization of this approach to dynamic features. The idea is that velocity in the image can be measured without the constraint of having an a priori knowledge of the scene. In this paper, a control law to position a camera mounted on the end effector of a robot, in such a way the image plane becomes parallel to a planar object, is presented. A dynamic visual servoing approach is used by defining a control loop upon the second order spatial derivatives of the optical flow. A fixating task, which guarantees the object stays in the camera field of view is joined to the first one. Then, results obtained on a 6 d.o.f. robot are laid out for the two studied tasks.

Abstract not found

Efficient real-time robotic tasks using a monocular vision system were previously developed with simple objects (e.g. white points on a black background), within a visual servoing context. Due to recent developments, it is now possible to design real-time visual tasks exploiting motion information in the image, estimated by robust algorithms. This paper proposes such an approach to track complex objects, such as a pedestrian. It consists in integrating the measured 2D motion of the object to recover its 2D-position in the image. The principle of the tracking task is to control the camera pan and tilt such that the estimated center of the object appears at the center of the image. Real-time experimental results demonstrate the efficiency and the robustness of the method. 1.

One major problem of underwater observation with an automatic engine is the instability of image acquisition and visualization. Indeed, small engines of this kind are subjected to low-frequency motions due to weak friction and water currents. In this paper, we propose to maintain stabilization in the image by controlling the pan and tilt motions of the camera attached to the engine, using techniques for target tracking by visual servoing. The main idea behind approach lies in the fact that, since it is very difficult to track a point in the images of an unknown and complex scene using geometrical tools, the position of a virtual point can be retrieved by the integration of its 2D motion. The motion estimation method we have used, called the RMR algorithm, provides the parameters of a selected motion model (for the task considered here, a constant one) and is perfectly suitable for real-time constraints and the complexity of an undersea image sequence. Our approach has been validated on a dry setup using two different sequences of underwater images. c ○ 2000 Academic Press Key Words: image stabilization; visual servoing; 2D motion; underwater vision. 1.

In this paper, a short survey of vision-based robot control (generally called visual servoing) is presented. Visual servoing concerns several field of research including vision systems, robotics and automatic control. Visual servoing can be useful for a wide range of applications and it can be used to control many different dynamic systems (manipulator arms, mobile robots, aircraft, etc.). Visual servoing systems are generally classified depending on the number of cameras, on the position of the camera with respect to the robot, on the design of the error function to minimize in order to reposition the robot. In this paper, we describe the main visual servoing approaches proposed in the literature. For simplicity, the examples in the survey focuses on manipulator arms with a single camera mounted on the end-effector. Examples are taken from work made at the University of Cambridge for the European Long Term Research Project VIGOR (Visually guided robots using uncalibrated cameras).

Visual servoing based upon geometrical features such as image points coordinates is now well set on. Nevertheless, this approach has the drawback that it usually needs visual marks on the observed object to retrieve geometric features. The idea developed here is that these features can be retrieved by integrating dynamic ones, which can be estimated without any a priori knowledge of the scene. Thus, more realistic objects can be used to achieve vision-based control such as tracking and fixation tasks. We detail control laws concerning these two tasks, first using integration of speed in the image and then by direct regulation of these dynamic parameters. Results are finally presented and comparisons are made between the two types of control methods. 1 Introduction The aim of visual servoing, as presented in [11, 13], is to control the robot displacements using visual features. One of the method used to complete such control laws is to apply the task function approach [19] to visual se...

This article reports on a reactivedocking behavior which uses a vision algorithm that grows linearly with the number of image pixels. The docking robot imprints (initializes) on a two-colored docking fiducial upon departing from the dock, then uses region statistics to adapt the color segmentation in changing lighting conditions. The docking behavior was implemented on a marsupial team of robots, where a daughter micro-rover had to re-enter the mother robot from an approach zone with a 2 meter radius and angular width with a tolerance of \Sigma5 and \Sigma2cm. Testing during outdoor conditions (noon, dusk) and challenging indoor scenarios (flashing lights) showed that using adaptation and imprinting was more robust than using imprinting alone.

This chapter is dedicated to a new way to achieve robotic tasks by 2D visual servoing. Contrary to most of related works in this domain where geometric visual features are usually used, we directly here consider the luminance of all pixels in the image. We call this new visual servoing scheme photometric visual servoing. The main advantage of this new approach is that it greatly simplifies the image processing required to track geometric visual features all along the camera motion or to match the initial visual features with the desired ones. However, as it is required in classical visual servoing, the computation of the so-called interaction matrix is required. In our case, this matrix links the time variation of the luminance to the camera motions. We will see that this computation is based on a illumination model able to describe complex luminance changes. However, since most of the classical control laws fail when considering the luminance as a visual feature, we turn the visual servoing problem into an optimization one leading to a new control law. Experimental results on positioning tasks validate the feasibility of photometric visual servoing and show its robustness regarding to approximated depths, Lambertian and