N. Papanikolopoulos, P. Khosla, and T. Kanade, "Vision and control techniques for robotic visual tracking," in In Proc. IEEE Int. Conf. Robotics and Autmation, 1991, vol. 1, pp. 851-- 856.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....might be initialized by taking the conjunction of the features in an image window that is known to contain the target image. However, computing moments of thresholded grayscale images is likely to be brittle to changes in lighting, point of view, etc. A more traditional approach is taken in Papanikolopoulos [1991], which tracks a set of correlation features on a rigid object, 48 but it is difficult to automaticMly select features that will provide robust correlation results [Thorpe, 1983] Swain [1990] introduces a fast scheme for recognizing and locating multi colored objects. However, since the ....

....in a fronto parallel plane, by translating the camera in a fronto parallel plane. The moments (which are known a priori) are computed on a binary image ob tained by thresholding the grayscale input image. This approach is likely to be brittle to changes in lighting, point of view, etc. Papanikolopoulos [1991] smoothly tracks pre selected features (image patches) It is not known how to select such features auto matically, although this problem has been and will doubtlessly continue to be a subject of investigation [Matthies et al. 1989; Thorpe, 1983] Waxman et al. 1988] saccadi cally track a set ....

N. Papanikolopoulos, P. Khosla, and T. Kanade, "Vi- sion and Control Techniques for Robotic Visual Tracking," In International Conference on Robotics and Automation. IEEE, 1991.

....allows to introduce the notion of hybrid task. This task is made of a combination of a primary task, which realizes the visual servoing, and a secondary task. The secondary task can be considered as a minimization of a cost function. The main applications of a secondary task are visual tracking [10], singularities and joints limits avoidance [7] The secondary task uses the unconstrained d.o.f. The velocities used in the secondary task are in open loop, under the constraint of a perfect achievement of the primary task. For the last couple of years, researches in the field of motion control ....

N. Papanikolopoulos, P.K. Khosla, and T. Kanade. Vision and control techniques for robotic visual tracking. In IEEE Int. Conf. on Robotics and Automation, pages 857-- 864, Sacramento, USA, April 1991.

....this approach to more complex features. 1 Introduction The use of visual information in robot control has been approached by several techniques. Many articles propose tracking algorithms based on the detection of moving objects in image space, and make a path planning with a Kalman filter [1][12][13] However, work dealing with the integration of visual information in a robot control loop shows two possible approaches. The first approach, Look and Move , has something in common with the articles quoted above. This is a technique where the position of the robot is estimated with the data ....

Papanikolopoulos N., P.K. Khosla, T. Kanade, Vision and Control techniques for robotic visual tracking, In Proceeding of the IEEE Conference on Robotics and Automation, April 1991, Sacramento, USA.

....discuss the incorporation of this method into the active contour framework where a shape subspace is used constrain shape variation. 1 Introduction Tracking objects with flexible shapes in video sequences is currently an important topic in the vision community. Methods include curve fitting [9], layered models [1, 2, 3] Bayesian reconstruction of 3 D models from video[6] and active contour models [10, 14, 15] Fitting curves to the outlines of objects has been attempted using various methods, including Snakes [8, 9] where an energy function is minimized so as to find the best fit. ....

....important topic in the vision community. Methods include curve fitting [9] layered models [1, 2, 3] Bayesian reconstruction of 3 D models from video[6] and active contour models [10, 14, 15] Fitting curves to the outlines of objects has been attempted using various methods, including Snakes [8, 9], where an energy function is minimized so as to find the best fit. As with other optimization methods, this approach suffers from local maxima. This problem is amplified when using real data where edge noise can prevent the fit of the contour to the desired object outline. In contrast, Blake et ....

N. Papanikolopoulos, P. Khosla, T. Kanade "Vision and Control Techniques for robotic visual tracking," In Proc. IEEE Int. Conf. Robotics and Autmation 1, 1991, pp. 851 -- 856.

....#d = #(qd ) be the reference features, then the visual servo problem is formulated as a potential minimization problem with the potential function being V (q) #d #(q) T (#d #(q) 1) B. Cont ol Law A typical control law is the steepestdk IO4kC law of (1) given by [1] 6] [7], 8] 9] q = J (#d #) 2) where J is thegeneralized inverse of Jand J isdk4KO by J = ## # c pc # c pc #q = # # J1 . Jn # # c Jr . 3) In thisdskBBBM4k c Jr is the robot Jacobi matrixexpressed in the camera coord MMM systemand J i is the image Jacobian for i th ....

N. Papanikolopoulos, P. K. Khosla, and T. Kanade, "Vision and control techniques for robotic visual tracking," in IEEE Int. Conf. Robotics and Automation, Sacramento, Calif., 1991, pp. 857--864.

....(the world coordinates) of the object to be manipulated. They used these results to generate movement commands for their servoing system. The results of their research were a body of techniques which required a precise calibration between the camera and world systems to insure adequate performance. [22, 21, 18, 26, 35] The idea of precise calibration between a mechanical device and photographic emulsions or photo electric sensors has been examined in great detail by researchers in non topographic photogrammetry. By using models which account for most of the aberration and lens defects in modern lenses, they ....

N. Papanikolopoulos, P. Khosla, and T. Kanade. Vision and control techniques for robotic visual tracking. In IEEE International Conference on Robotics and Automation, volume 1, pages 857--864, April 7-12 1991.

....devices are small parallel linkage manipulators used for moving single cameras. Examples of moveable eyes can be found in [17, 5] Eye in Hand In this configuration, a single camera is mounted on a robotic arm which provides movement capabilities. Examples of this configuration can be found in [1, 26]. Robotic Head Robotic heads are special purpose manipulators used to provide visual system capabilities similar to those found in biological systems. Most heads are binocular or trinocular. Examples of robotic heads can be found in [15, 25, 21, 8] Head in Hand In this configuration, a ....

N. Papanikolopoulos, P.K. Khosla, and T. Kanade. Vision and control techniques for robotic visual tracking. In Proceedings of the IEEE International Conference on Robotics and Automation, pages 857--864, Sacramento, California, April 1991.

....up to date trajectories from noisy and delayed outputs from different vision algorithms. Our previous work [4] coped with that problem in a similar way as in [39] using an ff Gamma fi Gamma fl filter, which is a form of a steady state Kalman filter. A similar approach can be found in papers by [34, 29, 6]. In [34] a sophisticated control scheme is described which combines a Kalman filter s estimation and filtering power with an optimal (LQG) controller which computes the robot s motion. The authors have presented good tracking results, as well as stated that the controller is robust enough so the ....

....from noisy and delayed outputs from different vision algorithms. Our previous work [4] coped with that problem in a similar way as in [39] using an ff Gamma fi Gamma fl filter, which is a form of a steady state Kalman filter. A similar approach can be found in papers by [34, 29, 6] In [34] a sophisticated control scheme is described which combines a Kalman filter s estimation and filtering power with an optimal (LQG) controller which computes the robot s motion. The authors have presented good tracking results, as well as stated that the controller is robust enough so the use of ....

N. Papanikolopoulos, T. Kanade, and P. Khosla. Vision and control techniques for robotic visual tracking. In Proceedings of the IEEE Conference on Robotics and Automation, 1991.

....et al. 5] Chaumette et al. 3] and Casta no et al. 2] These methods track feature points and effect servoing movements using an Image Jacobian which relates Cartesian movements with positional errors derived from the tracked features. Other methods include the work of Papanikopolous et al. [9], Koivo et al. 6] and Miller [12] The basic idea behind the Image Jacobian is to model the differential relationship between the camera system and the robotic control system in order to accurately predict the effects of small changes in one system on the other. It is a linear, position dependent ....

N. Papanikolopoulos, P. Khosla, and T. Kanade. Vision and control techniques for robotic visual tracking. In IEEE International Conference on Robotics and Automation, volume 1, pages 857--864, April 712 1991.

....that are tangent to constraint surfaces in the configuration space [29] One possible solution to this limitation is to use vision based techniques to control motion in the remaining directions. Thus, much research attention has recently been focused on vision based control (see, for example, [2, 3, 9, 10, 15, 16, 22, 27, 30, 31, 32, 33, 34, 35, 37]) Although vision based control has been used successfully for a number of tasks (for example, in welding applications [1, 5, 21] none of the systems referenced above lend themselves to task level specification of goals, and therefore, there are currently no automatic planning systems that can ....

N. Papanikolopoulos, P. K. Khosla, and T. Kanade. Vision and control techniques for robotic visual tracking. In Proc. IEEE Int'l Conference on Robotics and Automation, pages 857--864, April 1991.

....camera in a fronto parallel plane. The moments of the blob are computed on a binary image obtained by thresholding the greyscale input image. The values of the moments that identify the target are known a priori. This approach is likely to be brittle to changes in lighting, point of view, etc. Papanikolopoulos et al. 1991] smoothly tracks pre selected features (image patches) It is not known how to select such features automatically, although this problem has been and will doubtlessly continue to be a subject of investigation [Matthies et al. 1989; Thorpe, 1983] Waxman et al. 1988] saccadically track a set of ....

N. Papanikolopoulos, P. Khosla, and T. Kanade, "Vision and Control Techniques for Robotic Visual Tracking," In Proc. International Conference on Robotics and Automation. IEEE, 1991.

....probably, for the pragmatic reason that most robot controllers present an abstraction of the robot as a position controlled device. However some researchers in visual servoing who have built their own robot controllers at the axis level have all chosen to implement axis position control [9 11]. The redundant levels of control add to system complexity and may impact on closed loop performance [12] Exceptions to this approach are the citrus picking robot [13] which closes the vision loop about a hydraulic actu ator which is a natural velocity source, and also [14, 15] Weiss [2] ....

N. Papanikolopoulos, P. Khosla, and T. Kanade, "Vision and control techniques for robotic visual tracking," in Proc. IEEE Int. Conf. Robotics and Automation, pp. 857--864, 1991.

....of visual feedback. Many authors address the problem of servoing a camera to maintain a constant position relative to a known object with a single camera [5, 12, 16, 23, 28, 29] Wijesoma et al.: 27] describe a monocular hand eye system for planar positioning using image feedback. Other authors [21, 22] describe monocular visual servoing systems that use some type of adaptivity to compensate for unknown system parameters or to refine system calibration. Maru et al.: 18] describe algorithms for performing stereo vision based positioning. The principle differences between their approach and the ....

N. Papanikolopoulos, P. Khosla, and T. Kanade. Vision and control techniques for robotic visual tracking. In Proc. IEEE International Conference on Robotics and Automation, pages 857--864. IEEE Computer Society Press, 1991.

....to as the hand eye calibration. It has long been argued that proper use of visual measurements within a feedback loop can improve the accuracy and response of a hand eye system [3] Several authors have exhibited systems that employ visual feedback from a single end effector mounted camera [5, 13, 2]. However the use of only one camera places strong limitations on their capabilities. Systems employing feedback from stereo vision have been exhibited, but have focused on using reconstruction as the basis of the feedback system [1, 9] As with all position based systems, it is possible to ....

N. Papanikolopoulos, P.K. Khosla, and T. Kanade. Vision and control techniques for robotic visual tracking. In Proc. IEEE Int. Conf. on Robotics and Automation, pages 857--864, 1991.

....routine and visual state estimation in general. 1 Introduction A common characteristic of the visual servo control schemes reported to date is that the vision system is used in an outer command loop , which generates reference inputs to an inner robot control loop (e.g. 1] 3] 4] 5] [8], 12] This arrangement, which we shall refer to as a dual loop visual servo controller, is illustrated in Figure 1. In dual loop controllers, the vision loop typically runs at a frequency much lower than that of the robot controller. This difference in sampling rates is typically due to ....

N. Papanikolopoulos, P. K. Khosla, and T. Kanade. Vision and control techniques for robotic visual tracking. In Proc. IEEE Int'l Conference on Robotics and Automation, pages 857-- 864, April 1991.

No context found.

N. Papanikolopoulos, P.K. Khosla, and T. Kanade, "Vision and control techniques for robotic visual tracking," Proceedings of the IEEE International Conference on Robotics and Automation, 857-864, 1991.

....Real Time Software using Port Based Objects David B. Stewart Richard A. Volpe Pradeep K. Khosla CMU RI TR 93 11 Advanced Manipulators Laboratory, The Robotics Institute, and Department of Electrical and Computer Engineering Carnegie Mellon University 5000 Forbes Avenue Pittsburgh, PA 15213 July 1, 1993 I993 Carnegie Mellon Unviersity The research reported in this paper is supported, in part by, U.S. Army AMCOM and DARPA under contract DAAA 2189 C 0001, the National Aeronautics and Space Adw inislration (NASA) under contract NAGl 1075, the Depart ment of Electrica and ....

....9 Example of PID joint control . 9 Structure of state variable table mechanism for control module integration . 10 Example of module integration: Cartesian teleopemfion . 12 Generic framework of a port based object . 13 Example of visual servoing using inverse dynamics control module . 15 Example of visual servoing using damped least squares control module . 15 Example of combining modules: a computed torque controller . 17 Flowchart of the sender and receiver tasks for triple buffered ....

[Article contains additional citation context not shown here]

N. P. Papanikolopoulos, P. K. Khosla, and T. Kanade, "Vision and control techniques for robotic visual tracking", in Proc. of 1991.

....characteristics of the motion detection algorithm into a mathematical model for tracking. These differences allow the system to be used in environments that are inherently difficult to calibrate, such as underwater, in toxic environments, or in outer space. This paper extends our previous work [14][15][16] in Controlled Active Vision by allowing full 3D tracking of objects, by reducing the number of parameters that are estimated on line, and by presenting experimental results from tests performed using commercial manipulators. The experiments were performed on the Rapid Assembly System which ....

N.P. Papanikolopoulos, P.K. Khosla, and T. Kanade, "Vision and Control Techniques for Robotic Visual Tracking," in Proc. of the IEEE Int. Conf. on Robotics and Automation, pp. 857-864, April, 1991. page 17

No context found.

N. Papanikolopoulos, P. Khosla, and T. Kanade, "Vision and control techniques for robotic visual tracking," in In Proc. IEEE Int. Conf. Robotics and Autmation, 1991, vol. 1, pp. 851-- 856.

No context found.

N. Papanikolopoulos, P. K. Khosla, and T. Kanade, "Vision and control techniques for robotic visual tracking," in Proc. IEEE Int. Conf. Robotics and Automation, ICRA'91, Sacramento, CA, Apr. 1991, pp. 857--864.

No context found.

N. Papanikolopoulos, P.K. Khosla, and T. Kanada. Vision and control techniques for robotic visual tracking. In Proc. of the IEEE Int. Conf. on Robotics and Automation, volume 1, pages 857--864, 1991.

No context found.

N. Papanikolopoulos, P. K. Khosla, and T. Kanade, "Vision and control techniques for robotic visual tracking," in Proc. IEEE Int. Conf. Robotics and Automation, ICRA'91, Sacramento, CA, Apr. 1991, pp. 857--864.

No context found.

N. Papanikolopoulos, P.K. Khosla, and T. Kanade. Vision and control techniques for robotic visual tracking. In Proc. IEEE Int.Conf. Robotics and Automation, pp. 857--864, 1991.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC