Position Control Experiments Using Vision

This article presents results of experiments carried out to test the vision-based method of camera-space manipulation using a fully three-dimensional rigid-body mating task. In this work, the performance of the method under various operating conditions is studied experimentally using a large six-axis manipulator. Further, an analytical study is carried out to understand the influence of relative orientation of cameras and their number on the terminal positioning precision expected with the method.

Strategies for Increasing the Tracking Region of an Eye-in-Hand System by Singularity and Joint Limit Avoidance

An eye-in-hand system visually tracking objects can fail when the manipulator encounters a kinematic singularity or a joint limit. A solution to this problem is presented in which objects are visually tracked while the manipulator simultaneously avoids kinematic singularities and manipulator joint limits by moving in directions along which the tracking task space is unconstrained or redundant. A manipulability measure is introduced into the visual tracking objective function, providing an elegant and robust technique for deriving a control law that visually tracks objects while accounting for the configuration of the manipulator. Two different tracking strategies, one using a standard visual tracking strategy and the other using the newly proposed strategy, are experimentally compared on an actual hand/eye system. The experimental results demonstrate the effectiveness of the new method by showing that the tracking region of a manipulator tracking objects with planar motion can be greatly increased.

Visual Servoing Based on Image Motion

The general aim of visual servoing is to control the motion of a robot so that visual features acquired by a camera become superimposed with a desired visual pattern. Visual servoing based on geometrical features such as image point coordinates is now well established. Nevertheless, this approach has the drawback that it usually needs visual marks on the observed object to retrieve geometric features. The idea developed in this paper is to use motion in the image as the input of the control scheme since it can be estimated without any a priori knowledge of the observed scene. Thus, more realistic scenes or objects can be considered. Two different methods are presented. In the first method, geometric features are retrieved by integration of motion, which allows the use of classical control laws. This method is applied toa6degree-of-freedom positioning task. The authors show that, in such a case, an affine model of 2-D motion is insufficient to ensure convergence and that a quadratic model is needed. In the second method, the principle is to try to obtain a desired 2-D motion field in the image sequence. In usual image-based visual servoing, variations of visual features are linearly linked to the camera velocity. In this case, the corresponding relation is more complex, and the authors describe how it is possible to use this relation. This approach is illustrated with two tasks: positioning a camera parallel to a plane and following trajectory.

Trajectory estimation of a moving object using Kalman filter and Kohonen networks

A novel approach to estimate the real-time moving trajectory of an object is proposed in this paper. The object's position is obtained from the image data of a charge coupled device (CCD) camera, while a state estimator predicts the linear and angular velocities of the moving object. To overcome the uncertainties and noises residing in the input data, a Kalman filter and neural networks are utilized cooperatively. Since the Kalman filter needs to approximate a nonlinear system into a linear model in order to estimate the states, there still exist errors as well as uncertainties. To resolve this problem, in this approach, the Kohonen networks, which have a high adaptability to the memory of the input–output relationship, are utilized for the nonlinear region. In addition to this, the Kohonen network, as a sort of neural network, can effectively adapt to the dynamic variations and become robust against noises. This approach is derived from the observation that the Kohonen network is a type of a self-organized map and is spatially oriented, which makes it suitable for determining the trajectories of moving objects. The superiority of the proposed algorithm compared with the extended Kalman filter is demonstrated through real experiments.

Nonlinear visual mapping model for 3-D visual tracking with uncalibrated eye-in-hand robotic system

A new control scheme for uncalibrated robotic visual tracking problem is proposed that compromises the computational expenses of overall system with offline modeling and online control. A nonlinear visual mapping model for the uncalibrated hand-eye coordination is first proposed with an artificial neural network implementation. An online visual tracking controller is then developed together with a real-time motion planner. To improve the system performance, the control scheme is also integrated with a feedforward controller to compensate unknown object motions. Extensive simulations and experiments demonstrate the effectiveness of the proposed control scheme.